Final Report Summary - CHEMOGUIDE (Chemotherapy guidance based on efficacy of treatment regimen in cancer cells from the individual patient)
Executive Summary:
The primary aim of the ChemoGuide Demonstration project was to clinically validate a prototype technology developed in a previous “Research for SMEs” project, ICSC. The ChemoGuide project consists of three previous partners (2CX, BBH and MDL) and two new partners: Universitäts Klinik Eppendorf Hamburg (UKE) and University of Colorado Cancer Center (UCCC). An efficient integration and alignment of interest of the new partners has been established through several face-to-face meetings in Copenhagen, Hamburg, Bilbao and Denver.
The strengthening of the previous ICSC project group with these two clinical partners in different territories has made it clear that there is a tremendous medical need for a technology like the ChemoGuide that open for individualizing the medical treatment of not only colorectal cancer patients; but for most types of late stage solid cancers.
During the ChemoGuide project the prototype technology was optimized for clinical use by moving the cell screening format from 2D to 3D. This allows production of micro-tumors ex-vivo from the primary tumor tissue obtained from individual patients. There is clear evidence in the scientific literature that such 3D micro tumors resemble the in-vivo situation better than the standard 2D cell system used for culturing cell lines. In the ChemoGuide project we have established standard operating procedures for handling tissue biopsies of patient tumors from colon and from patient liver metastasis. The 3D micro-tumors have been compared to the original tumor/metastasis of the patient using standard pathological characterization.
Further, the material composition and the layout of the disposable array was optimized for the 3D cell-screening format.
The full technology has been tested in tumor tissue obtained from more than 230 colorectal cancer patients. A clinical trial including the two clinical centers in Copenhagen and Hamburg was launched in the summer of 2014. 102 colorectal cancer patients (stage IIb, III and IV) have until now been enrolled in a prospective, non-interventional clinical trial. The full correlation between ChemoGuide screening results and patient outcome (Progression-Free-Survival) will be finalized two years after enrolment of the last patient.
At University of Colorado Cancer Center (Denver, USA) we have positively evaluated the technology in Patient-Derived-Tumor-Xenografs (PDTX).
The ChemoGuide project will have significant impact on the following six topics: 3D tumoroid based functional testing show value in predicting patient outcome after specific chemotherapy treatment; The ChemoGuide technology is applicable to other cancers than colorectal cancer; Dissemination of results; Regulatory requirements defined; Basis for commercial benefit to the involved SMEs; Basis for commercial benefit to the European society.
Project Context and Objectives:
ChemoGuide will take the former “Research for SMEs” project, ICSC, through a clinical study to prepare the technology for final commercialization. The ICSC has developed an in-vitro test that allows identification of new drug candidates by conducting high throughput compound synthesis and screening in colon cancer cells isolated from individual patients. Furthermore, the technology has been developed to include not only large numbers of single compounds but also compounds in combinations as used in cancer therapy. Up to 4 compounds at a time have been combined and tested in tumor cells freshly isolated from patients. Such screening technology has opened up for unique possibilities of testing patient’s cancer cells against existing chemo compounds for the aim of identifying the most efficacious therapy (often combination therapy) from approved cancer drugs based on prospective functional resistance and efficacy in cancer cells from the individual patient.
Turning the ICSC project towards the chemo therapy optimization (ChemoGuide test) was pushed by the surgeons and oncologists involved in the ICSC project, as there is a huge and unmet need for such kind of testing, before starting a chemo therapy. The ICSC technology has proven to be highly suitable for such testing as it allows numerous tests to be done on very little amount of tissue. The technology is formatted as an easy-to-use bench top technology based on disposable chips, which allows the test to be done locally at the hospital. Thus, the ChemoGuide product will be a routine test for identification of resistance and efficacy to all standard treatment regimens for the individual colorectal cancer patient. The product will be distributed to the customer as arrays with treatment regimens ready for routine testing on-site and readable in a microarray scanner.
The technology benefits from a former technology developed within 2cureX called Chemocellomics, which integrates chemical synthesis or immobilization of drug molecules and cell attachment and growth for the aim of conducting high throughput cell-based screening in a micro-format that allows the screening to be done directly in primary cells from patients. This is accomplished using a polymer optimized for chemical synthesis or immobilization and at the same time constitutes optimal environment for cell attachment and growth. ChemoCellomics® is the first technology that merges chemistry and cell-based screening on the same solid support and the technology is protected by issued patents.
Objectives:
The aim of the ChemoGuide project is to optimize an ICSC prototype technology for clinical use, and to conduct a prospective, non-interventional clinical study to prove that there is a significant correlation between the result of the ChemoGuide test and the actual patient outcome in order to mature the technology for subsequent commercialization.
The project addresses colorectal cancer (CRC), which is among the most diagnosed cancers in the developed world with almost 300,000 and 150,000 new annual cases in the EU and the USA, respectively. Matching the individual patient with a specific medical treatment regimen carries a tremendous potential of both improving treatment quality and cost effectiveness. This is especially the case in the area of cancer, as it is burdened by the lowest median drug efficacy of all therapeutic areas.
The main objective of the ChemoGuide project is to evaluate the ChemoGuide test for its ability to predict drug resistance and efficacy of the prevailing standard treatment regimens in 3D micro-tumors from stage IIb, III and IV colorectal cancer patients. The clinical study will be conducted at Bispebjerg Hospital (BBH) in Copenhagen, Denmark and at Universitäts Klinik Eppendorf Hamburg (UKE) in Hamburg, Germany. Treatment response will be evaluated with CT-scans six, twelve and twenty-four months after initiation of chemotherapy. The results of the test will be evaluated by monitoring the Progression-Free-Survival of the individual patients. A parallel study will be launched at the Cancer Center at University of Colorado where colorectal tumor tissue from patients is xenograf’ed into immune compromised mice (Patient-Derived-Tumor-Xenograf, PDTX). Tumor tissue from the PDTX mice will be evaluated for its ability to produce 3D micro-tumors.
Project Results:
Work progress and Science & Technology (S&T) results of the ChemoGuide project:
The overall aim of the ChemoGuide project is to optimize an ICSC prototype technology for clinical use, and to conduct a prospective, non-interventional clinical study to prove that there is a significant correlation between the results of the ChemoGuide test and the actual patient outcome. A successful clinical trial will mature the technology for subsequent commercialization.
The ChemoGuide project has achieved all major objectives (Milestones) laid out at the launch of the project. Further, an early evaluation of the technology in live tissue from colorectal cancer patients allowed the ChemoGuide project to optimize the test for the final clinical trial that was launched in the summer of 2014.
Milestones and Deliverables are presented in Table 1 and Table 2 (Figures attachment).
The S&T results of the ChemoGuide project is organized by presenting a short description of the planned work from DoW, a summary of the results achieved and summaries of each deliverable under the corresponding Work package (WP).
WP1: Project management
Short description of planned work from DoW:
A consortium agreement will be written and executed between all partners and EC before signing of the Grant Agreement.
A small and efficient management team will be established consisting of the lead participants from each of the two SMEs. The coordinator (2CX) will be the chairman. The management team will be responsible for the overall strategic and execution of the project plan. The management will instigate procedures for scientific and financial reporting to be used by all partners. Each partner will be responsible for reporting this information to the management on a quarterly basis. The coordinator (chairman of the management team) will assemble all information received from participants and prepare a semi-annual report for EC.
Summary of WP results:
The Consortium Agreement was prepared and executed. Ethical approvals for conducting the proposed clinical trial obtained in Denmark and Germany.
Deliverable 1.1 Ethical clearance:
The ethical clearances in Denmark and Germany allow us to take tissue obtained through surgical or biopsical procedure from patient with colorectal cancer. The patient will have to provide a written consent before tissue sampling. The Ethical approvals allowed us broadly enroll all colorectal cancer patients that receive chemotherapy. The approval presented the patients in three sub protocols (see below). In the final clinical trial we decided to concentrate our effort on sub protocol 1 (stage IIb and III patients) and sub protocol 2b (stage IV – liver resected).
The patients are stratified into the following three groups:
A. Sub protocol 1 (stage IIb-III) includes colorectal cancer patients without distant metastases at the time of diagnosis.
B. Sub protocol 2a (stage IV – no liver resection) includes patients suffering from metastatic colon cancer (stage IV) subject to chemotherapy.
C. Sub protocol 2b (stage IV – liver resection) includes patients suffering from metastatic colon cancer (stage IV) with hepatic metastases that are subject to surgical removal.
Patient enrolment:
Inclusion criteria:
1. Adenocarcinoma in colon or rectum verified by biopsy – or observed by colonoscopy, where the tumor has not been verified by biopsy
2. No presence of distant metastases determined by CT- or PET-CT scan (sub protocol 1). Presence of liver metastasis determined by CT- or PET-CT scan (sub protocol 2b)
3. Indication for elective surgery
4. Age ≥ 18 years
5. The patient understands Danish/German
6. The patient is capable
Exclusion criteria:
1. Rectal cancer requiring radiation therapy prior to surgery
2. Evaluation of removed colorectal tumor tissue does not show any signs of cancer or shows that the cancer is not an adenocarcinoma
3. Intraoperatively identified metastases in e.g. liver, mesentery, or peritoneum
Deliverable 1.2 Signed Consortium Agreement:
The ChemoGuide partners agreed to a Consortium Agreement that is in full agreement with the rules outlined EU Commission.
WP2: Validation and correction of prototype array:
Short description of planned work from DoW:
Several prototype arrays were developed in the ICSC project with the common important requirement being that they should be compatible with a standard DNA micro array reader. The reason behind this was that the implementation of the ICSC technology should not require huge investments in equipment. Many departments at universities and hospitals already have such DNA array readers convincing us that this should be the format.
Beside the economic and practical considerations, the form factor of the technology allowed the array to be quite small (compared to standard plate reader format). The array design of interest developed for this project is an array consisting of 48 small wells for holding the polymer in which the test compounds are immobilized.
Production of the ChemoGuide prototype arrays which are needed for the ChemoGuide project. The arrays need to be verified with regard to:
• Flatness of the array and the reservoir
• Identification of suitable bio-compatible, water resistant adhesive.
Upon completion of the patient screening (WP4), the array prototype will go through a final optimisation process based on the experiences gathered during the screening activities. The ChemoGuide array is made ready for mass production.
Summary of WP results:
An initial prototype arrays was prepared and used in the initial testing of patient material. During the test it became apparent that several properties (see Deliverable 2.1 and 2.2) had to be improved. Several iteration with more than 2000 arrays produced resulted in an array that possess the properties listed in the DoW and a high degree of transparency that allow high quality imaging.
Deliverable 2.1 Delivery of prototype arrays needed for the ChemoGuide project:
The disposable array had to have three key characteristics: a) high degree of flatness of the bottom allowing imaging-based read-out without continues focusing: b) highly transparent bottom to allow high quality imaging and c) made in a biocompatible material.
During the optimization process (described in detail in the uploaded Deliverable 2.1) we discovered after having achieved the above presented specifications that the arrays during inject moulding process became afflicted with welding lines that disturbed our automated image analysis.
Removal of the welding lines was achieved by polishing the mould. This was a very time consuming and resource demanding task. In the subsequent Deliverable 2.2 this problem was solved by switching to a different array material.
The in Deliverable 2.1 produced proto-type arrays were used in our initial studies on patient material (see WP 4). The initial transfer of patient tissue to arrays for production of 3D micro tumors and the optimization of the properties of the arrays occurred in parallel. In bi-weekly Skype meetings we coordinated this biological and technical optimization.
Deliverable 2.2 Final optimizations of prototype array:
In parallel to the initial biological tests conducted in the disposable array described above a method was developed that allowed large-scale production. Following these procedures, over 2000 arrays have been produced. Up to now, the fabrication of ChemoGuide arrays has been done by inject moulding in PMMA. Looking forward to mass production, we have improved the techniques used for the mould manufacturing introducing techniques that obtain reproducible arrays from mould to mould. The material Polystyrene was chosen for the final mass-produced arrays. Further, other changes were introduced such as using opaque polymers to improve the light reflection in order to optimize the array reading and results obtained.
The arrays design and manufacturing has evolved during the project, as some challenges has been needed to overcome, to obtain a highly reproducible array manufactured with the right technique, compatible with the desired functions of the disposable array. During the project over two thousand units has been manufactured, during the different iterations to adapt the array to a mass production compatible array that fulfill the requirements (flatness, transparency, and biocompatibility) (Details presented in the uploaded Deliverable 2.2). A long path has led to an array design that fulfills the project expectations.
WP 3: Setting up test sites at 3 locations
Short description of planned work from DoW:
BBH, UCCH and UCCC to be equipped with appropriate instrumentation for launching the clinical trial. Thus standard equipment for handling the tissue and subsequent cell culturing is available being LAF benches (sterile flow hood), CO2 incubators and microscopes.
A DNA array scanner suitable for the testing is available. It is though expected that a low cost dedicated fluorescence reader (presently being developed in collaboration between 2CX and the Danish Technological Institution, DELTA) will be ready for implementation before starting the present project.
A final “package" of standard procedures (SOP’s) will be prepared including:
• Inclusion of patients
• Sampling from the tumor
• Handling of tumor tissue
• Carry out of the testing
Necessary ethical approvals will be obtained from Ethics Committee in all three territories. Collection and storage of patient information will be approved by the appropriate Register of Data Protection.
Summary of WP results:
All three test sites were established according to plan. Further, the necessary ethical and data protection approvals were obtained.
Deliverable 3.1 Test site ready for trial; Copenhagen:
Laboratory facility for conducting the clinical testing in Denmark is setup at:
Bispebjerg Hospital
Bispebjerg Bakke 23
Building 8, 2. Tværvej
DK-2400 Copenhagen NV
Denmark
The test facility is located at the University Hospital Bispebjerg (BBH) in close proximity to the department of surgery where tumor resection is taking place.
The test facility that is ready for the clinical trial is a standard cell culture facility with laminar flow benches, CO2-incubators, centrifuges, inverted microscopes and standard cell culture utensils. Micro-tumor growth properties are monitored using a 3D imaging scanner, oCelloscope (BioCell, Philips).
At the outset of the ChemoGuide project we had planned to conduct the testing of different treatment regimens in 2D cell cultures. However early in the project it became clear that moving to 3D would create micro-tumors that much more closely resemble the tumor in the patient.
A standard operating procedure (SOP) has been developed for handling the tissue and for preparation of the micro-tumors (detailed description presented in the uploaded Deliverable 3.1).
The micro-tumors are challenged with the combination therapies: FOLFOX, FOLFIRI and the individual drugs (5-fluoruracil, leucovorin, oxaliplatin and SN38) for seven days. Growth of the micro-tumors, or inhibition thereof, is monitored daily (see Fig. 1 and 2).
Deliverable 3.2.: Test site ready for clinical trial in Hamburg:
Laboratory facility for conducting the clinical testing in Germany is setup at:
Lab for GI Oncology and Response Prediction
Department of Oncology and Hematology
University of Hamburg
University Hospital Hamburg - Eppendorf
Martinistrasse 52
D-20246 Hamburg
The laboratory facility is localized in the Main Science Building (see map) centrally at the University Medical Center Hamburg Eppendorf (UKE).
The equipment and test procedures used at UKE is a copy of the setup in Copenhagen (see Deliverable 3.1 for details). In order for us to compare clinical data obtained at the two sites it is very important that all aspects of the clinical testing are identical.
Employees from BBH and 2curex (2CX) have travelled to Hamburg on average every second month to ensure an efficient transfer of knowhow, and to ensure best practice. Another important aspect of the screening operation is that all pathological and biochemical data of the individual patient are uploaded to a common database.
The database is constructed after input from all clinical partners and approved by the Danish and German data protection authorities.
Deliverable 3.3.: Test site ready for clinical trial in Denver:
Laboratory facility for conducting the clinical testing in the US is setup at:
University of Colorado
School of Medicine
Division of Medical Oncology
1250 14th St.
Denver, CO 80217
USA
The Messersmith laboratory is a fully equipped cancer laboratory localized at the main campus in direct proximity to the patient clinic.
The equipment and test procedures used at UCCC are a copy of the setup in Copenhagen (see Deliverable 3.1 for details). In order for us to compare clinical data obtained at the two sites it is very important that all aspects of the clinical testing are identical.
We decided to conduct the actual human clinical trial exclusively in Europe. Conducting the trial in Copenhagen and Hamburg allowed easy exchange of researchers between the two “neighboring” sites during the clinical trial. In the US additional experiments were undertaken where human patient tissue was xenografed into immunedeficient mice and tissue from these animals was examined using the ChemoGuide technology (see Deliverable 4.3).
WP 4: Testing of colon cancer cells from 100 patients for sensitivity towards standard treatment regimen
Short description of planned work from DoW:
The ChemoGuide technology will be evaluated in a clinical setting where 50 -80 patients will be enrolled at two oncology centers: BBH (Denmark) and UCCH (Germany). The testing will be executed by personnel trained to conduct such screening using the ChemoGuide test. The microarray devices distributed to the oncology centers will hold standard combination regimens used in EU and USA. The ratio between the individual drugs will be varied and each treatment regimen will be run in triplicate. The trial will be used to evaluate the ability of the test to predict drug resistance and efficacy to the prevailing standard treatment regimens in stage III and stage IV CRC patients.
UCCC has established a colorectal patient-derived-tumor-xenograft (PDTX) model in immune-compromised mice. Tissue will be taken from the patient xenografts and ChemoGuide screening will be conducted as done in the human clinical trial (D.4.1). The PDTX model opens for tailoring treatment of the individual mouse carrying a specific patient’s tumor. Such prospective clinical study can first be conducted in humans after a successful finalization of above described clinical trial.
Summary of WP results:
In June 2014 we began enrolling patients at the two screening sites for the ChemoGuide project. In total 235 patients have been enrolled. Of these patients 133 were not included in the ChemoGuide test optimization and screening activities. Exclusion of the 133 patients was mainly due to a too low number of produced 3D micro-tumors, necrosis or infections. In the later period of the project (last 6 months – since January 2015) the loss of patient tissue that could not be used for screening dropped from 57% (133 out of 253 patients) to 14% (19 out of 134 patients). The decision to move from 2D cell culture to 3D micro-tumors (presented in the RP1 report) has proven to be the right decision as we and others have now shown that such 3D cultures closely resembles the tumor in-vivo in both pathological, histological and functional properties.
It was possible to establish 3D micro-tumors from tumor tissue obtained from PDTX mice. The growth behavior of these micro-tumors resembles was observed when taking tissue directly from patients. A limited number of PDTX mice were treated with Irinotecan and show close to complete response. Micro-tumors prepared from these mice likewise showed sensitivity to the active metabolite of Irinotecan, SN38. Additional studies are needed to complete a comprehensive correlation study.
Deliverable 4.1: >150 patients screened:
In June 2014 we initiated the clinical screening where 3D micro-tumors from patients were screened against standard regimens used in Europe and the US (Fig. 4.1.1).
Image analysis and data presentation is described in Deliverable 5.1.
Patient enrolment and screening optimization:
In total, we have enrolled 235 patients in the ChemoGuide project. The patients can be divided into the following groups:
A. 67 patients included in the ChemoGuide clinical study. From these patients 112 arrays were prepared and screened.
B. 133 patients were excluded due to inability to produce 3D micro-tumor cultures, infections or necrotic tissue.
C. 35 patients were used to optimize the test. From these patients 142 arrays were prepared and screened. Optimization was necessary due to a large variation in growth behavior of the 3D micro-tumors.
Test results and patient information has been databased for all 253 enrolled patients. The patient information including treatment outcome is databased in Denmark and Germany for each their national patients
Group A: The ChemoGuide clinical study.
Table 4.1.2 shows 102 patients were included in the ChemoGuide screening activities of these 67 patients were part of the clinical trial whereas 35 patients where used to tackle a problem of too large variation in the first prototype array (Group C).
Group B: Excluded due to infection or inability to produce 3D micro-tumor cultures.
After the launch of the ChemoGuide project, we decided to move the screening format from 2D cell culture to 3D micro-tumors. This was a major undertaking; but we are in no doubt that, we took the right decision as we and others have now clearly shown a clear resemblance between the 3D cultures and the tumor in-vivo with regard to cellular characteristics and functional response. This change of format did however result in a significant loss (133 of 253) of tissue samples that did not produce 3D micro-tumors that could be used in the screening activity. Fortunately, the loss of successful preparations has gone down from 57% (133 out of 253 patients) to 14% (19 out of 134 patients) when looking at enrolled patient since January of 2015. A success rate of 86% is acceptable in the clinical setting.
Group C: Too large variation in screening results observed in the first prototype array.
After having screened a number of patients both in Copenhagen and in Hamburg we observed a too large a variation in the growth rates of the 3D micro-tumors.
Several patients showed a nice dose-response of the tested treatment regimens when measuring inhibition of growth after 7 days; but in between we observed huge variation in both controls and treatment regimens (see Fig. 4.1.3).
To identify the problem causing the large variability of our screens we worked through the procedures for tissue handling, drug loading into arrays, buffer composition etc. We finally concluded that variation had to come from the arrays. The material used for the inject moulding of the first prototype was Poly (methyl Methacrylate) (PMMA). We decided to test different materials [COP (Cyclic Olefin Polymer), COC (cyclic olefin copolymer and PS (Polystyrene)]. We were able to produce COC and COP arrays with the existing mould; but PS required production of a new mould (see Deliverable 2.2).
Several growth experiments in PS showed that this material provides an array system with low variability (Fig. 4.1.4). The arrays used for the clinical screening are therefore made of PS.
Deliverable 4.2: >150 patients screened:
A database system has been established that holds patient information and screening results. The Database complies with the Danish and German data laws. Screening results and associated analysis and treatment recommendation for all enrolled patients are up-loaded to database. Patient information on the German patients is however kept in Germany and the merger of screening results with patient information will be conducted by the German partners in Hamburg.
Overall description of the database:
• The database servers are physically located at Bispebjerg Hospital (Copenhagen)
• All access to the database requires a user specific username and password
• All communication between the user (the webbrowser) and the database is certificate based 128bit TLS encrypted
• Identifying information for Danish patients (CPR number) is 256bit encrypted and stored on a separate server located in a separate room from the server which contains the rest of the database information
• Identifying information for German patients is kept at Universitaetsklinikum Hamburg-Eppendorf (Hamburg)
• Screening data for both Danish and German patients are stored in the database at Bispebjerg Hospital (Copenhagen)
Data on all enrolled patients (see Deliverable 5.1) has been up-loaded into the database.
Database upload:
Conducting clinical studies in two different countries require a strict control of all procedures from tissue recovery, handling, testing, data analysis and data storage. These procedures need to be identical for the two test sites, and at the same time we need to accommodate the different national legislation. On the latter point both sites stored all test results in the same database located in Copenhagen. However personal information on the German patients was stored at Universitaetsklinikum Hamburg-Eppendorf in Hamburg. The merger of test results and patient information/treatment outcome on Danish and German patients is conducted in Denmark and in Germany, respectively.
Test preparations are uploaded into the Database using the form presented in Fig. 4.2.1.
Deliverable 4.3: Correlation between screening data and treatment outcome in PDTX:
It was possible to establish 3D micro-tumors from tumor tissue obtained from PDTX mice. The growth behavior of these micro-tumors resembles was observed when taking tissue directly from patients. A limited number of PDTX mice were treated with Irinotecan and show close to complete response. Micro-tumors prepared from these mice likewise showed sensitivity to the active metabolite of Irinotecan, SN38. Additional studies are needed to complete a comprehensive correlation study.
Preparation of 3D micro-tumors from PDTX mice and correlation to treatment efficacy in PDTX mice:
Colorectal xenograft samples were obtained from PDTXs previously established at the University of Colorado Cancer Center, Denver, CO, USA. Briefly, fresh colorectal tumor tissue was obtained from consenting patients at the University of Colorado Hospital in accordance with protocols approved by the Colorado Multiple Institutional Review Board. Tumor samples were cut into 3 mm3 pieces and implanted into the flank of 4-6 weeks old female athymic nuce mice. When tumor volume reached 1000-1500 mm3, the tumor was excised and passaged into subsequent generations. For micro-tumor preparation, tumors from untreated mice were excised, placed in cold PBS with antibiotics and transported to the laboratory on ice. All animal experiments were performed under an approved protocol by the Institutional Animal Care and Use Committee.
We obtained a 86% success rate (12 out of 14 tested) in generating micro-tumors from PDTX mice.
Three animals were treated and showed almost complete response (70-80% remission) when treated with Irinotecan. Tissue from tumors of these animals were tested for treatment sensitivity in the ChemoGuide test (Fig. 4.3.1).
A low number of animals tested did not allow a correlation analysis to be made; but it could be concluded that with the ChemoGuide technology it is possible to produce micro-tumors from the PDTX mice, and that it was observed that patient micro-tumors were sensitive to Irinotecan as observed in the treated mice.
WP 5: Evaluation of the correlation between test results and patient out-come
Short description of planned work from DoW:
As part of the standard protocol for patients receiving adjuvant chemotherapy treatment for colorectal cancer the treatment efficacy will be evaluated with CT-scans three and six months after initiation of chemotherapy.
The RECIST response criteria will be used to evaluate the disease response or progression following the chemotherapy. The result of a ChemoGuide screening is a predication of tumor resistance and sensitivity towards chemotherapy for the individual patient. By comparing the individual patient's ChemoGuide screening prediction and the RECIST classification as monitored by CT-scanning, it will be possible to verify the prognostic value of the ChemoGuide screening system. Importantly significant correlation will demonstrate that the ChemoGuide screening system is highly useful for prospective chemo-sensitivity screening.
At the completion of the patient screening (WP4) the treatment efficacy data from each individual patient will be collected, this includes CT-scans, the RECIST evaluation and other medical evaluations performed by the oncologists.
Once the patient data has been collected, a correlation between the individual patient’s treatment efficacy and the ChemoGuide screening results will be performed. ChemoGuide predictions of both sensitivity and resistance towards the treatment regimens will be used in this correlation.
Summary of WP results:
In total we enrolled 253 patients in the ChemoGuide project. Of these 102 patients where included in the clinical study. The screening results have been analyzed using image analysis where the growth of the 3D micro-tumors have been measured and categorized in three sensitivity groups: No response (resistance); Partial response; Complete response.
The full correlation between screening results and patient outcome will be conducted when all stage IV and stage IIB/III patients have been evaluated using the RECIST and the Disease-Free-Survival (DFS) criteria, respectively.
In the original DoW we anticipated that at least 50 patients would be enrolled in Denmark, Germany and the US. This has, as presented in the approved amendment, been changed such that all clinical studies involving human patients will be conducted in Denmark and Germany. This change has not influenced the total number of patients enrolled in the project. In the laboratory of our US-partner we have instead expanded the ChemoGuide project by including experiments with Patient-Derived-Tumor-Xenografs in mice. These new experiments have resulted in addition of a new deliverable: 4.3 Correlation between screening data and treatment outcome in PDTX.
In DoW we presented that the screening results would be correlated to the RECIST criteria monitored using CT scanning 3 and 6 months after launch of adjuvant treatment. This will still be the case for stage IV patients with liver metastasis however; it cannot be done in the stage IIb/III patients as all patients we received had got their primary tumor completely resected. For this patient group we will correlate the screening results with Disease-Free-Survival (DSF). As this patient group has an overall survival is significantly better than the stage IV patients (see Fig. 5.1.1) we will need to wait to conduct the final correlation between test results and patient outcome until two years after launch of the adjuvant treatment for the last enrolled patient. For stage IV patients this correlation can be conducted two years earlier.
Deliverable 5.1: Screening data evaluated and correlation to patient outcome concluded:
Screening process and data acquisition:
The clinical screening in the ChemoGuide project is conducted using a 3D image scanning reader (oCelloscope). As mentioned in RP1 we decided early in the ChemoGuide project to move from a 2D cell format to 3D. This decision led us to investigate a new image-based reader system that allows fast detection of 3D information in bright field illumination. Using a novel scanning technology, the reader is capable of collecting the 3D information in a single scan.
The screening process is shown schematically in Fig. 5.1.2.
After setting up the 3D micro-tumors in the arrays we follow the growth for seven days with daily scans. In the final product we may lower the number of scans as the amount of data collected is very high (approx. 35 Gb per patient).
Important for this test is that the 3D micro-tumors are perturbed as little as possible. We have therefore omitted use of molecular probes for monitoring metabolic functions, and conducted an end-point measure of growth inhibition evoked by the different treatment regimens.
The image analysis is conducted using a semi-automatic image analysis algorithm developed in collaboration with Philips BioCell who as developed the 3D reader (oCelloscope, http://www.philips.com/content/corporate/en_AA/biocell/home.html).
Data analysis:
Images acquired by the oCelloScope are analyzed in a two-step process. Frist an automated algorithm stiches the images together and generates z-stacks, it then identifies 3D micro-tumors using edge finding, contrast based methods while also incorporating the 3D information in the acquired images.
A secondary program is used to manually check the results of the automated algorithm and is used to remove false positives and add 3D micro-tumors that were not correctly identified using alternative micro-tumor-finding algorithms. When all micro-tumors are identified their 2D areas are calculated. (Fig. 5.1.3).
The area is monitored daily over seven days (Fig. 5.1.4).
From Fig. 5.1.4 it can be seen that the growth rate varies from patient to patient. This is a reflection of the patient heterogeneity that not only is observed with regard to the sensitivity/resistance to certain treatment regimens, it is also observed when monitoring the growth potential of the patient’s tumor. It is therefore very important that each patient is its own control when evaluating the effectiveness of the different treatment regimens. This is exemplified in Fig. 5.1.5 where the responsiveness of three patients towards Folfox and Folfiri (with and without Cetuximab) is monitored.
From Fig. 5.1.5 it is observed that Pt601 show low sensitivity to all four treatment regimens, whereas Pt606 show moderate sensitivity to Folfox with no extra benefit of including Cetuximab. The last Pt574 is a highly sensitive patient where efficacy is observed for all four treatments with noticeable improvement of including Cetuximab.
Using the data presentation shown in Fig. 5.1.4 the responsiveness of the individual patient to each treatment regimen will be grouped in: No response (resistance); Partial response; Complete response.
This grouping will be correlated to the RECIST (stage IV patients) and the Disease-free-survival (stage IIb/III patients) (see below).
Correlation of screening results with patient outcome:
In total we have enrolled 253 patients in the ChemoGuide project. The enrolment started in June 2014. Of these patients, we successfully prepared 3D micro-tumors from 102 patients and included these in the screening activity. The reason for the relatively large number of patients that could not be used for screening is explained in Deliverable 4.1. On average we have been able to run two arrays on each patient (in total 254 arrays). All most half (46%) of the included patients had metastatic disease (stage IV with liver metastasis) (Deliverable 4.1 Table 4.1.1).
The results of the clinical screening conducted in the ChemoGuide project will not at this early stage influence which treatment regimen is offered to the patients. The purpose of the study is to show whether there is a correlation between the screening results and the responsiveness of the patient to the given treatment. The screening results will be associated with clinical staging and associated biomedical data of the individual patient. The screening results for the patients with metastatic disease (stage IV with liver metastasis) will be correlated to the RECIST evaluation. The RECIST measurements will be based on CT scans performed prior to the initiation of chemotherapy and 3 and 6 months after the initiation of chemotherapy. In short, the RECIST evaluation will measure lesion diameters and their development, the result of the RECIST evaluation will be a division of patients into four categories depending on their response to the chemotherapy treatment:
• Complete Response (CR): Disappearance of all target lesions. Any pathological lymph nodes must have reduction in short axis to <10 mm.
• Partial Response (PR): At least a 30% decrease in the sum of diameters of target lesions, taking as reference the baseline sum diameters.
• Stable Disease (SD): Neither sufficient shrinkage to qualify for PR nor sufficient increase to qualify for PD, taking as reference the smallest sum diameters since the treatment started
• Progressive Disease (PD): At least a 20% increase in the sum of diameters of target lesions, taking as reference the smallest sum on study (this includes the baseline sum if that is the smallest on study). In addition to the relative increase of 20%, the sum must also demonstrate an absolute increase of at least 5 mm.
For the stage IIB/III patients (54%) the RECIST criteria cannot be used when evaluating the treatment efficiency as all these patients have got their primary tumor completely resected. The clinical endpoint to be used for these patients will be Disease-free-survival monitored two years after launch of the adjuvant treatment.
For both patient categories we are not allowed (due to the blinded design of the study) to conduct the correlation between screening results and patient outcome before all stage IV patients have been through the 6 month CT scanning and the stage IIb/III patients have been followed for two years after start of treatment.
Example of how to correlate screening results to RECIST evaluation is presented in Fig. 5.1.6.
WP 6: Dissemination
Short description of planned work from DoW:
The oncologists of the ChemoGuide project will present the test findings for colleagues at their own clinical center and at corresponding center in their respective countries. Further it is planned to present the findings at international meetings organized by e.g. the European Cancer Organisation and the American Society of Clinical Oncology.
The commercialization of the ChemoGuide test will have to be supported by strong clinical evidence. The best way scientifically to present that is through peer reviewed international journals.
The clinical partners of the present project represent the end-users of the final test. The commercialization of the ChemoGuide test will therefore initially occur at these test centers and similar sites in Europe and USA. To support this commercialization, we will prepare descriptive material that presents the scientific/clinical concept behind the test and how the test results are going to support the tailoring of individualized adjuvant treatment.
This material will include semi-scientific reports, flyers and a web-site.
Summary of WP results:
The technology and the clinical findings (see Deliverable 5.1) have been presented to oncologists at the two partnering hospitals (Bispebjerg Hospital and Universitaetsklinikum Hamburg-Eppendorf) and other hospitals in the Copenhagen and Hamburg region. This has increased the number of hospitals involved from the present two to seven. Even though a complete correlation analysis between screening results and patient outcome is not yet possible (see Deliverable 5.1) the tremendous interest clearly shows a belief in the technology and clinical unmet need.
Discussions are ongoing with further oncology centers in the participating countries and in Sweden.
Deliverable 6.1: Presentation of clinical finding
Presentation of clinical findings:
The presentation and subsequent commercialization of the ChemoGuide screening test will occur through the established test centers in Denmark and Germany. To increase the knowledge of this new screening technology in the clinical environment we have several meetings with oncology colleagues at the partnering hospitals and beyond. The information presented has been very positively received and resulted in continue collaboration between the ChemoGuide partners and inclusion of additional hospitals in both Germany and Denmark (Hvidovre Hospital and Hillerød Hospital) that will recruit further patients for the ChemoGuide analysis.
Technical and clinical findings have been discussed among the ChemoGuide project members in numerous Skype meetings and several face-to-face meetings (2. April 2014; 16. April 2014; 12-13. May 2014; 28-31. July 2014; 22-23. Oct. 2014; 19. Dec 2014; 14. Jan 2015; 9-10. Febr 2015; 1. April 2015; 15-16. April 2015; 9. June 2015). This very active interaction between the partners has been key to the success of this multidisciplinary project.
In the ChemoGuide project we have prioritized presenting the technology to clinical and pharmaceutical institutions to support a future commercialization rather than writing scientific publications. The resemblance between 3D micro-tumors developed from liver metastasis and the lesion in-vivo has however, been presented at the American Association of Cancer Research 2014 meeting (San Diego, 4-11 April 2014; see Fig. 6.1.1).
Deliverable 6.2: Commercial product description
The partners behind the ChemoGuide project will continue (after finalization of the EU-funded project) the collaboration towards commercialization of the ChemoGuide test. Laboratory facilities at Bispebjerg Hospital (Copenhagen) and at Universitaetsklinikum Hamburg-Eppendorf (Hamburg) will thus continue to conduct follow-up evaluation of the enrolled patients to evaluate the predictive power of the test. MicroLiquid (Bilbao) will continue the production of the latest test arrays.
Commercial promotion of the test has been conducted latest at the BIO conference in June 2015 Philadelphia (USA). A semi-scientific description of the technical/clinical concept behind the test has been prepared through a Teaser and a Slide deck (see Fig. 6.2.1).
The web-site for 2cureX has been updated to include data from the ChemoGuide project.
Potential Impact:
The impact of the ChemoGuide project is separated into the following five topics:
1. 3D tumoroid based functional testing show value in predicting patient outcome after specific chemotherapy treatment
2. The ChemoGuide technology is applicable to other cancers than colorectal cancer
3. Dissemination of results
4. Regulatory requirements defined
5. Basis for commercial benefit to the involved SMEs
6. Basis for commercial benefit to the European society
Ad 1: 3D tumoroid based functional testing show value in predicting patient outcome after specific chemotherapy treatment.
In total we enrolled 253 patients in the ChemoGuide project. Of these 102 patients where included in the clinical study. The screening results have been analyzed using image analysis where the growth of the 3D micro-tumors have been measured and categorized in three sensitivity groups: No response (resistance); Partial response; Complete response.
The prospective, non-interventional clinical study conducted in the ChemoGuide project is designed such that all used treatment regimens are included in the screening array; but the result of the ChemoGuide test does not influence the selection of treatment. The patients received standard treatment as proposed by the oncologist and when patient outcome is available it will be investigated whether the ChemoGuide test correctly predicted outcome. Preliminary observations clearly suggest that such positive correlation exists (see Deliverable 5.1). Even though full analysis of the predictiveness of the test cannot be revealed earlier than 2-3 years after enrolment of the last patient in the study, oncologists from the partnering cancer centers designed an interventional study on stage IV metastatic colorectal cancer patients where the metastasis (most often liver) cannot be resected. This group constitutes the largest proportion of the stage IV patients, and the patients having the poorest prognosis.
The proposed study will be conducted at the partnering hospitals in Denmark and Germany plus one of the largest cancer centers in the United Kingdom. It is a major endorsement of the ChemoGuide test that the oncologists this early will embark on using the test in designing the treatment regimen for the individual patient.
The impact for the patients will be significant as the stage IV colorectal cancer patients have very little time left for selecting an efficacious treatment. Colorectal cancer is one of the most diagnosed cancers in Europe. Alone in the three countries where the intervention study will be conducted more than 20.000 colorectal cancer stage IV patients are diagnosed yearly.
Not only will a fast selection of an efficacious treatment benefit the patient, it will also result in major cost savings for the health care systems.
In the original DoW for the ChemoGuide project we described how genomic profiling had become synonymous with Personalized medicine: “The concept of individualized treatment – personalised medicine – was introduced as a conceptual part of the genome project more than two decades ago. The ability to design an individualized treatment therefore became synonymous with correlating the patient’s genetic profile to his/her’s drug responsiveness. Genotyping of patients has proven successful in identification of responsive patients to single targeted drugs like Herceptin (HER-2/neu expression) and Erbitux/Tarceva (KRAS or BRAF mutation). However, this has not been the case when trying to match specific combination therapies to individual patients.”
Since the launch of the ChemoGuide project the notion that genomic information is insufficient in tailoring the treatment of the individual patient has received significant support (for recent publications see Le Tourneau et al. (2015) Molecularly targeted therapy based on tumor molecular profiling versus conventional therapy for advanced cancer (SHIVA): a multicentre, open-label, proof-of-concept, randomised, controlled phase 2 trial. The Lancet 16, 1324-34; Friedman et al. (2015) Precision medicine for cancer with next-generation functional diagnostics. Nature Rev Cancer 15, 747-756).
The need and impact for clinically validated functional test will increase dramatically in the coming years. Here we see the test developed and clinically tested in the ChemoGuide to be in the forefront of cell-functional tests.
Ad 2: The ChemoGuide technology is applicable to other cancers than colorectal cancer
In the ChemoGuide project we have concentrated our effort on colorectal cancer. However, a subsequent introduction of the technology to other types of cancer is important both a commercial and societal point of view.
During the ChemoGuide project other cancers like pancreas, liver, breast and head & neck were tested for our ability to prepare 3D micro-tumors. It all cases it turned out positive. In conclusion it is likely that the preparation of 3D micro-tumors developed in ChemoGuide is applicable to other solid tumors.
It is obvious that expanding the test into other solid cancer will greatly increase the impact of the ChemoGuide test.
Ad 3: Dissemination of results
The ChemoGuide project has a clear commercial objective of providing a test for tailoring the medical treatment of colorectal cancer patients. It has therefore been the focus of the project group to protect developments and finding by patent filings. Subsequently the project will pursue dissemination through scientific publications.
One patent was filed in 2014 by 2cureX. Hagel and Thastrup: Identifying compounds modifying a cellular phenotype. PCT/DK2015/050197. The discoveries behind the patent is solely made by 2cureX and related to the integration of technologies for testing of patient material and for patient diagnostics. The patent is within the competences and business strategy of 2cureX. 2cureX is therefore the sole owner of the patent in accordance with the Consortium Agreement:
“8.1 Foreground ownership
RTD Provider Foreground (e.g. Intellectual Property) developed during the course and falling within the Field shall be vested to and owned by one or more of the SMEs. Transfer of rights in RTD provider Foreground to one or more of the SMEs shall be granted on Fair and Reasonable conditions. RTD Provider Foreground which is Patient Material shall at all times vest in RTD Provider and shall principally not be transferable, unless specified by a separate Agreement. The distribution of the RTD Provider Foreground between the SMEs is to be decided by Executive Board, along the protocol described below. It will be the responsibility, including the financial obligation, of the relevant SME(s) to file and prosecute associated patent applications. The relevant SME(s) will pay close attention to who the actual inventors are. The RTD and the associated inventors will be responsible for delivering all material necessary for the patent filing.
Foreground ownership shall be determined along the following principles, or as, exceptionally, otherwise agreed by the Executive Board:
2CX shall receive Foreground that is directly related to the integration of technologies for testing of patient material and for patient diagnostics and 2CX shall receive Foreground related to testing of patient material;
MLD shall receive Foreground that is directly related to the manufacturing of polymer devices and associated injection moulding procedures.
In case of conflicts with regard to the distribution of the ownership the Executive Board will decide.
The relevant SME has the obligation including bearing the associated cost of protecting the Foreground. In case no SME wishes to comply with this obligation the ownership shall remain with the relevant RTD provider.”
Ad 4: Regulatory requirements defined
The regulatory requirements needed for commercialization of the ChemoGuide test is the following:
• Establishment of an EN ISO 13485 compliant quality management system (QMS)
• Obtaining the certification of the EN ISO 13485 QMS by a relevant certification body (TÜV Süd)
• Establishment of the required technical documentation for the ChemoGuide test
• Ensure concordance with the IVD directive 98/79/EC
• Presentation of the clinical validation results to demonstrate the benefit of ChemoGuide test for the patient and the health care system
• Issuing a declaration of conformity for the ChemoGuide test
2cureX has identified the company, Simply Quality – Qualitätsmanagementberatung in Weilheim, Germany to manage the process of getting above certifications obtained.
Ad 5: Basis for commercial benefit to the involved SMEs
Two SMEs (2cureX and MLD) we partners in the ChemoGuide project. MLD has been responsible for optimization, prototyping and mass production of the disposable arrays used in the ChemoGuide test whereas 2cureX has assembled the ChemoGuide arrays with the treatment regimens and distributed them to the clinical partners for testing.
The successful production of the disposable arrays by MLD opens for commercial benefit to the SME by being the future mass producer of the arrays. Expansion into other types of cancers will require production of a large number of arrays. Further the knowledge of producing arrays for 3D cell cultures will become valuable as research tool in other biological and therapeutic areas.
2cureX is planning a subsequent interventional study with clinical partners in Denmark, Germany and the United Kingdom. The ChemoGuide project has paved the way for conduction of this final clinical study before commercialization into the clinical environment. Important for the subsequent roll-out of the test in major European cancer clinics are that the already involved partners are key opinion leaders in each their country.
It is essential for the commercial success of the test that the clinical partners of the ChemoGuide project has agreed to present the predictive power of the test to other major oncology centers in Europe.
2cureX will provide the integrated test with disposable arrays containing treatment regimens and test procedure. Data analysis can be conducted on-site or 2cureX can provide the service and as part thereof compare the results of the individual test with data from a steadily expanding database.
List of Websites:
www.chemoguide.eu
The primary aim of the ChemoGuide Demonstration project was to clinically validate a prototype technology developed in a previous “Research for SMEs” project, ICSC. The ChemoGuide project consists of three previous partners (2CX, BBH and MDL) and two new partners: Universitäts Klinik Eppendorf Hamburg (UKE) and University of Colorado Cancer Center (UCCC). An efficient integration and alignment of interest of the new partners has been established through several face-to-face meetings in Copenhagen, Hamburg, Bilbao and Denver.
The strengthening of the previous ICSC project group with these two clinical partners in different territories has made it clear that there is a tremendous medical need for a technology like the ChemoGuide that open for individualizing the medical treatment of not only colorectal cancer patients; but for most types of late stage solid cancers.
During the ChemoGuide project the prototype technology was optimized for clinical use by moving the cell screening format from 2D to 3D. This allows production of micro-tumors ex-vivo from the primary tumor tissue obtained from individual patients. There is clear evidence in the scientific literature that such 3D micro tumors resemble the in-vivo situation better than the standard 2D cell system used for culturing cell lines. In the ChemoGuide project we have established standard operating procedures for handling tissue biopsies of patient tumors from colon and from patient liver metastasis. The 3D micro-tumors have been compared to the original tumor/metastasis of the patient using standard pathological characterization.
Further, the material composition and the layout of the disposable array was optimized for the 3D cell-screening format.
The full technology has been tested in tumor tissue obtained from more than 230 colorectal cancer patients. A clinical trial including the two clinical centers in Copenhagen and Hamburg was launched in the summer of 2014. 102 colorectal cancer patients (stage IIb, III and IV) have until now been enrolled in a prospective, non-interventional clinical trial. The full correlation between ChemoGuide screening results and patient outcome (Progression-Free-Survival) will be finalized two years after enrolment of the last patient.
At University of Colorado Cancer Center (Denver, USA) we have positively evaluated the technology in Patient-Derived-Tumor-Xenografs (PDTX).
The ChemoGuide project will have significant impact on the following six topics: 3D tumoroid based functional testing show value in predicting patient outcome after specific chemotherapy treatment; The ChemoGuide technology is applicable to other cancers than colorectal cancer; Dissemination of results; Regulatory requirements defined; Basis for commercial benefit to the involved SMEs; Basis for commercial benefit to the European society.
Project Context and Objectives:
ChemoGuide will take the former “Research for SMEs” project, ICSC, through a clinical study to prepare the technology for final commercialization. The ICSC has developed an in-vitro test that allows identification of new drug candidates by conducting high throughput compound synthesis and screening in colon cancer cells isolated from individual patients. Furthermore, the technology has been developed to include not only large numbers of single compounds but also compounds in combinations as used in cancer therapy. Up to 4 compounds at a time have been combined and tested in tumor cells freshly isolated from patients. Such screening technology has opened up for unique possibilities of testing patient’s cancer cells against existing chemo compounds for the aim of identifying the most efficacious therapy (often combination therapy) from approved cancer drugs based on prospective functional resistance and efficacy in cancer cells from the individual patient.
Turning the ICSC project towards the chemo therapy optimization (ChemoGuide test) was pushed by the surgeons and oncologists involved in the ICSC project, as there is a huge and unmet need for such kind of testing, before starting a chemo therapy. The ICSC technology has proven to be highly suitable for such testing as it allows numerous tests to be done on very little amount of tissue. The technology is formatted as an easy-to-use bench top technology based on disposable chips, which allows the test to be done locally at the hospital. Thus, the ChemoGuide product will be a routine test for identification of resistance and efficacy to all standard treatment regimens for the individual colorectal cancer patient. The product will be distributed to the customer as arrays with treatment regimens ready for routine testing on-site and readable in a microarray scanner.
The technology benefits from a former technology developed within 2cureX called Chemocellomics, which integrates chemical synthesis or immobilization of drug molecules and cell attachment and growth for the aim of conducting high throughput cell-based screening in a micro-format that allows the screening to be done directly in primary cells from patients. This is accomplished using a polymer optimized for chemical synthesis or immobilization and at the same time constitutes optimal environment for cell attachment and growth. ChemoCellomics® is the first technology that merges chemistry and cell-based screening on the same solid support and the technology is protected by issued patents.
Objectives:
The aim of the ChemoGuide project is to optimize an ICSC prototype technology for clinical use, and to conduct a prospective, non-interventional clinical study to prove that there is a significant correlation between the result of the ChemoGuide test and the actual patient outcome in order to mature the technology for subsequent commercialization.
The project addresses colorectal cancer (CRC), which is among the most diagnosed cancers in the developed world with almost 300,000 and 150,000 new annual cases in the EU and the USA, respectively. Matching the individual patient with a specific medical treatment regimen carries a tremendous potential of both improving treatment quality and cost effectiveness. This is especially the case in the area of cancer, as it is burdened by the lowest median drug efficacy of all therapeutic areas.
The main objective of the ChemoGuide project is to evaluate the ChemoGuide test for its ability to predict drug resistance and efficacy of the prevailing standard treatment regimens in 3D micro-tumors from stage IIb, III and IV colorectal cancer patients. The clinical study will be conducted at Bispebjerg Hospital (BBH) in Copenhagen, Denmark and at Universitäts Klinik Eppendorf Hamburg (UKE) in Hamburg, Germany. Treatment response will be evaluated with CT-scans six, twelve and twenty-four months after initiation of chemotherapy. The results of the test will be evaluated by monitoring the Progression-Free-Survival of the individual patients. A parallel study will be launched at the Cancer Center at University of Colorado where colorectal tumor tissue from patients is xenograf’ed into immune compromised mice (Patient-Derived-Tumor-Xenograf, PDTX). Tumor tissue from the PDTX mice will be evaluated for its ability to produce 3D micro-tumors.
Project Results:
Work progress and Science & Technology (S&T) results of the ChemoGuide project:
The overall aim of the ChemoGuide project is to optimize an ICSC prototype technology for clinical use, and to conduct a prospective, non-interventional clinical study to prove that there is a significant correlation between the results of the ChemoGuide test and the actual patient outcome. A successful clinical trial will mature the technology for subsequent commercialization.
The ChemoGuide project has achieved all major objectives (Milestones) laid out at the launch of the project. Further, an early evaluation of the technology in live tissue from colorectal cancer patients allowed the ChemoGuide project to optimize the test for the final clinical trial that was launched in the summer of 2014.
Milestones and Deliverables are presented in Table 1 and Table 2 (Figures attachment).
The S&T results of the ChemoGuide project is organized by presenting a short description of the planned work from DoW, a summary of the results achieved and summaries of each deliverable under the corresponding Work package (WP).
WP1: Project management
Short description of planned work from DoW:
A consortium agreement will be written and executed between all partners and EC before signing of the Grant Agreement.
A small and efficient management team will be established consisting of the lead participants from each of the two SMEs. The coordinator (2CX) will be the chairman. The management team will be responsible for the overall strategic and execution of the project plan. The management will instigate procedures for scientific and financial reporting to be used by all partners. Each partner will be responsible for reporting this information to the management on a quarterly basis. The coordinator (chairman of the management team) will assemble all information received from participants and prepare a semi-annual report for EC.
Summary of WP results:
The Consortium Agreement was prepared and executed. Ethical approvals for conducting the proposed clinical trial obtained in Denmark and Germany.
Deliverable 1.1 Ethical clearance:
The ethical clearances in Denmark and Germany allow us to take tissue obtained through surgical or biopsical procedure from patient with colorectal cancer. The patient will have to provide a written consent before tissue sampling. The Ethical approvals allowed us broadly enroll all colorectal cancer patients that receive chemotherapy. The approval presented the patients in three sub protocols (see below). In the final clinical trial we decided to concentrate our effort on sub protocol 1 (stage IIb and III patients) and sub protocol 2b (stage IV – liver resected).
The patients are stratified into the following three groups:
A. Sub protocol 1 (stage IIb-III) includes colorectal cancer patients without distant metastases at the time of diagnosis.
B. Sub protocol 2a (stage IV – no liver resection) includes patients suffering from metastatic colon cancer (stage IV) subject to chemotherapy.
C. Sub protocol 2b (stage IV – liver resection) includes patients suffering from metastatic colon cancer (stage IV) with hepatic metastases that are subject to surgical removal.
Patient enrolment:
Inclusion criteria:
1. Adenocarcinoma in colon or rectum verified by biopsy – or observed by colonoscopy, where the tumor has not been verified by biopsy
2. No presence of distant metastases determined by CT- or PET-CT scan (sub protocol 1). Presence of liver metastasis determined by CT- or PET-CT scan (sub protocol 2b)
3. Indication for elective surgery
4. Age ≥ 18 years
5. The patient understands Danish/German
6. The patient is capable
Exclusion criteria:
1. Rectal cancer requiring radiation therapy prior to surgery
2. Evaluation of removed colorectal tumor tissue does not show any signs of cancer or shows that the cancer is not an adenocarcinoma
3. Intraoperatively identified metastases in e.g. liver, mesentery, or peritoneum
Deliverable 1.2 Signed Consortium Agreement:
The ChemoGuide partners agreed to a Consortium Agreement that is in full agreement with the rules outlined EU Commission.
WP2: Validation and correction of prototype array:
Short description of planned work from DoW:
Several prototype arrays were developed in the ICSC project with the common important requirement being that they should be compatible with a standard DNA micro array reader. The reason behind this was that the implementation of the ICSC technology should not require huge investments in equipment. Many departments at universities and hospitals already have such DNA array readers convincing us that this should be the format.
Beside the economic and practical considerations, the form factor of the technology allowed the array to be quite small (compared to standard plate reader format). The array design of interest developed for this project is an array consisting of 48 small wells for holding the polymer in which the test compounds are immobilized.
Production of the ChemoGuide prototype arrays which are needed for the ChemoGuide project. The arrays need to be verified with regard to:
• Flatness of the array and the reservoir
• Identification of suitable bio-compatible, water resistant adhesive.
Upon completion of the patient screening (WP4), the array prototype will go through a final optimisation process based on the experiences gathered during the screening activities. The ChemoGuide array is made ready for mass production.
Summary of WP results:
An initial prototype arrays was prepared and used in the initial testing of patient material. During the test it became apparent that several properties (see Deliverable 2.1 and 2.2) had to be improved. Several iteration with more than 2000 arrays produced resulted in an array that possess the properties listed in the DoW and a high degree of transparency that allow high quality imaging.
Deliverable 2.1 Delivery of prototype arrays needed for the ChemoGuide project:
The disposable array had to have three key characteristics: a) high degree of flatness of the bottom allowing imaging-based read-out without continues focusing: b) highly transparent bottom to allow high quality imaging and c) made in a biocompatible material.
During the optimization process (described in detail in the uploaded Deliverable 2.1) we discovered after having achieved the above presented specifications that the arrays during inject moulding process became afflicted with welding lines that disturbed our automated image analysis.
Removal of the welding lines was achieved by polishing the mould. This was a very time consuming and resource demanding task. In the subsequent Deliverable 2.2 this problem was solved by switching to a different array material.
The in Deliverable 2.1 produced proto-type arrays were used in our initial studies on patient material (see WP 4). The initial transfer of patient tissue to arrays for production of 3D micro tumors and the optimization of the properties of the arrays occurred in parallel. In bi-weekly Skype meetings we coordinated this biological and technical optimization.
Deliverable 2.2 Final optimizations of prototype array:
In parallel to the initial biological tests conducted in the disposable array described above a method was developed that allowed large-scale production. Following these procedures, over 2000 arrays have been produced. Up to now, the fabrication of ChemoGuide arrays has been done by inject moulding in PMMA. Looking forward to mass production, we have improved the techniques used for the mould manufacturing introducing techniques that obtain reproducible arrays from mould to mould. The material Polystyrene was chosen for the final mass-produced arrays. Further, other changes were introduced such as using opaque polymers to improve the light reflection in order to optimize the array reading and results obtained.
The arrays design and manufacturing has evolved during the project, as some challenges has been needed to overcome, to obtain a highly reproducible array manufactured with the right technique, compatible with the desired functions of the disposable array. During the project over two thousand units has been manufactured, during the different iterations to adapt the array to a mass production compatible array that fulfill the requirements (flatness, transparency, and biocompatibility) (Details presented in the uploaded Deliverable 2.2). A long path has led to an array design that fulfills the project expectations.
WP 3: Setting up test sites at 3 locations
Short description of planned work from DoW:
BBH, UCCH and UCCC to be equipped with appropriate instrumentation for launching the clinical trial. Thus standard equipment for handling the tissue and subsequent cell culturing is available being LAF benches (sterile flow hood), CO2 incubators and microscopes.
A DNA array scanner suitable for the testing is available. It is though expected that a low cost dedicated fluorescence reader (presently being developed in collaboration between 2CX and the Danish Technological Institution, DELTA) will be ready for implementation before starting the present project.
A final “package" of standard procedures (SOP’s) will be prepared including:
• Inclusion of patients
• Sampling from the tumor
• Handling of tumor tissue
• Carry out of the testing
Necessary ethical approvals will be obtained from Ethics Committee in all three territories. Collection and storage of patient information will be approved by the appropriate Register of Data Protection.
Summary of WP results:
All three test sites were established according to plan. Further, the necessary ethical and data protection approvals were obtained.
Deliverable 3.1 Test site ready for trial; Copenhagen:
Laboratory facility for conducting the clinical testing in Denmark is setup at:
Bispebjerg Hospital
Bispebjerg Bakke 23
Building 8, 2. Tværvej
DK-2400 Copenhagen NV
Denmark
The test facility is located at the University Hospital Bispebjerg (BBH) in close proximity to the department of surgery where tumor resection is taking place.
The test facility that is ready for the clinical trial is a standard cell culture facility with laminar flow benches, CO2-incubators, centrifuges, inverted microscopes and standard cell culture utensils. Micro-tumor growth properties are monitored using a 3D imaging scanner, oCelloscope (BioCell, Philips).
At the outset of the ChemoGuide project we had planned to conduct the testing of different treatment regimens in 2D cell cultures. However early in the project it became clear that moving to 3D would create micro-tumors that much more closely resemble the tumor in the patient.
A standard operating procedure (SOP) has been developed for handling the tissue and for preparation of the micro-tumors (detailed description presented in the uploaded Deliverable 3.1).
The micro-tumors are challenged with the combination therapies: FOLFOX, FOLFIRI and the individual drugs (5-fluoruracil, leucovorin, oxaliplatin and SN38) for seven days. Growth of the micro-tumors, or inhibition thereof, is monitored daily (see Fig. 1 and 2).
Deliverable 3.2.: Test site ready for clinical trial in Hamburg:
Laboratory facility for conducting the clinical testing in Germany is setup at:
Lab for GI Oncology and Response Prediction
Department of Oncology and Hematology
University of Hamburg
University Hospital Hamburg - Eppendorf
Martinistrasse 52
D-20246 Hamburg
The laboratory facility is localized in the Main Science Building (see map) centrally at the University Medical Center Hamburg Eppendorf (UKE).
The equipment and test procedures used at UKE is a copy of the setup in Copenhagen (see Deliverable 3.1 for details). In order for us to compare clinical data obtained at the two sites it is very important that all aspects of the clinical testing are identical.
Employees from BBH and 2curex (2CX) have travelled to Hamburg on average every second month to ensure an efficient transfer of knowhow, and to ensure best practice. Another important aspect of the screening operation is that all pathological and biochemical data of the individual patient are uploaded to a common database.
The database is constructed after input from all clinical partners and approved by the Danish and German data protection authorities.
Deliverable 3.3.: Test site ready for clinical trial in Denver:
Laboratory facility for conducting the clinical testing in the US is setup at:
University of Colorado
School of Medicine
Division of Medical Oncology
1250 14th St.
Denver, CO 80217
USA
The Messersmith laboratory is a fully equipped cancer laboratory localized at the main campus in direct proximity to the patient clinic.
The equipment and test procedures used at UCCC are a copy of the setup in Copenhagen (see Deliverable 3.1 for details). In order for us to compare clinical data obtained at the two sites it is very important that all aspects of the clinical testing are identical.
We decided to conduct the actual human clinical trial exclusively in Europe. Conducting the trial in Copenhagen and Hamburg allowed easy exchange of researchers between the two “neighboring” sites during the clinical trial. In the US additional experiments were undertaken where human patient tissue was xenografed into immunedeficient mice and tissue from these animals was examined using the ChemoGuide technology (see Deliverable 4.3).
WP 4: Testing of colon cancer cells from 100 patients for sensitivity towards standard treatment regimen
Short description of planned work from DoW:
The ChemoGuide technology will be evaluated in a clinical setting where 50 -80 patients will be enrolled at two oncology centers: BBH (Denmark) and UCCH (Germany). The testing will be executed by personnel trained to conduct such screening using the ChemoGuide test. The microarray devices distributed to the oncology centers will hold standard combination regimens used in EU and USA. The ratio between the individual drugs will be varied and each treatment regimen will be run in triplicate. The trial will be used to evaluate the ability of the test to predict drug resistance and efficacy to the prevailing standard treatment regimens in stage III and stage IV CRC patients.
UCCC has established a colorectal patient-derived-tumor-xenograft (PDTX) model in immune-compromised mice. Tissue will be taken from the patient xenografts and ChemoGuide screening will be conducted as done in the human clinical trial (D.4.1). The PDTX model opens for tailoring treatment of the individual mouse carrying a specific patient’s tumor. Such prospective clinical study can first be conducted in humans after a successful finalization of above described clinical trial.
Summary of WP results:
In June 2014 we began enrolling patients at the two screening sites for the ChemoGuide project. In total 235 patients have been enrolled. Of these patients 133 were not included in the ChemoGuide test optimization and screening activities. Exclusion of the 133 patients was mainly due to a too low number of produced 3D micro-tumors, necrosis or infections. In the later period of the project (last 6 months – since January 2015) the loss of patient tissue that could not be used for screening dropped from 57% (133 out of 253 patients) to 14% (19 out of 134 patients). The decision to move from 2D cell culture to 3D micro-tumors (presented in the RP1 report) has proven to be the right decision as we and others have now shown that such 3D cultures closely resembles the tumor in-vivo in both pathological, histological and functional properties.
It was possible to establish 3D micro-tumors from tumor tissue obtained from PDTX mice. The growth behavior of these micro-tumors resembles was observed when taking tissue directly from patients. A limited number of PDTX mice were treated with Irinotecan and show close to complete response. Micro-tumors prepared from these mice likewise showed sensitivity to the active metabolite of Irinotecan, SN38. Additional studies are needed to complete a comprehensive correlation study.
Deliverable 4.1: >150 patients screened:
In June 2014 we initiated the clinical screening where 3D micro-tumors from patients were screened against standard regimens used in Europe and the US (Fig. 4.1.1).
Image analysis and data presentation is described in Deliverable 5.1.
Patient enrolment and screening optimization:
In total, we have enrolled 235 patients in the ChemoGuide project. The patients can be divided into the following groups:
A. 67 patients included in the ChemoGuide clinical study. From these patients 112 arrays were prepared and screened.
B. 133 patients were excluded due to inability to produce 3D micro-tumor cultures, infections or necrotic tissue.
C. 35 patients were used to optimize the test. From these patients 142 arrays were prepared and screened. Optimization was necessary due to a large variation in growth behavior of the 3D micro-tumors.
Test results and patient information has been databased for all 253 enrolled patients. The patient information including treatment outcome is databased in Denmark and Germany for each their national patients
Group A: The ChemoGuide clinical study.
Table 4.1.2 shows 102 patients were included in the ChemoGuide screening activities of these 67 patients were part of the clinical trial whereas 35 patients where used to tackle a problem of too large variation in the first prototype array (Group C).
Group B: Excluded due to infection or inability to produce 3D micro-tumor cultures.
After the launch of the ChemoGuide project, we decided to move the screening format from 2D cell culture to 3D micro-tumors. This was a major undertaking; but we are in no doubt that, we took the right decision as we and others have now clearly shown a clear resemblance between the 3D cultures and the tumor in-vivo with regard to cellular characteristics and functional response. This change of format did however result in a significant loss (133 of 253) of tissue samples that did not produce 3D micro-tumors that could be used in the screening activity. Fortunately, the loss of successful preparations has gone down from 57% (133 out of 253 patients) to 14% (19 out of 134 patients) when looking at enrolled patient since January of 2015. A success rate of 86% is acceptable in the clinical setting.
Group C: Too large variation in screening results observed in the first prototype array.
After having screened a number of patients both in Copenhagen and in Hamburg we observed a too large a variation in the growth rates of the 3D micro-tumors.
Several patients showed a nice dose-response of the tested treatment regimens when measuring inhibition of growth after 7 days; but in between we observed huge variation in both controls and treatment regimens (see Fig. 4.1.3).
To identify the problem causing the large variability of our screens we worked through the procedures for tissue handling, drug loading into arrays, buffer composition etc. We finally concluded that variation had to come from the arrays. The material used for the inject moulding of the first prototype was Poly (methyl Methacrylate) (PMMA). We decided to test different materials [COP (Cyclic Olefin Polymer), COC (cyclic olefin copolymer and PS (Polystyrene)]. We were able to produce COC and COP arrays with the existing mould; but PS required production of a new mould (see Deliverable 2.2).
Several growth experiments in PS showed that this material provides an array system with low variability (Fig. 4.1.4). The arrays used for the clinical screening are therefore made of PS.
Deliverable 4.2: >150 patients screened:
A database system has been established that holds patient information and screening results. The Database complies with the Danish and German data laws. Screening results and associated analysis and treatment recommendation for all enrolled patients are up-loaded to database. Patient information on the German patients is however kept in Germany and the merger of screening results with patient information will be conducted by the German partners in Hamburg.
Overall description of the database:
• The database servers are physically located at Bispebjerg Hospital (Copenhagen)
• All access to the database requires a user specific username and password
• All communication between the user (the webbrowser) and the database is certificate based 128bit TLS encrypted
• Identifying information for Danish patients (CPR number) is 256bit encrypted and stored on a separate server located in a separate room from the server which contains the rest of the database information
• Identifying information for German patients is kept at Universitaetsklinikum Hamburg-Eppendorf (Hamburg)
• Screening data for both Danish and German patients are stored in the database at Bispebjerg Hospital (Copenhagen)
Data on all enrolled patients (see Deliverable 5.1) has been up-loaded into the database.
Database upload:
Conducting clinical studies in two different countries require a strict control of all procedures from tissue recovery, handling, testing, data analysis and data storage. These procedures need to be identical for the two test sites, and at the same time we need to accommodate the different national legislation. On the latter point both sites stored all test results in the same database located in Copenhagen. However personal information on the German patients was stored at Universitaetsklinikum Hamburg-Eppendorf in Hamburg. The merger of test results and patient information/treatment outcome on Danish and German patients is conducted in Denmark and in Germany, respectively.
Test preparations are uploaded into the Database using the form presented in Fig. 4.2.1.
Deliverable 4.3: Correlation between screening data and treatment outcome in PDTX:
It was possible to establish 3D micro-tumors from tumor tissue obtained from PDTX mice. The growth behavior of these micro-tumors resembles was observed when taking tissue directly from patients. A limited number of PDTX mice were treated with Irinotecan and show close to complete response. Micro-tumors prepared from these mice likewise showed sensitivity to the active metabolite of Irinotecan, SN38. Additional studies are needed to complete a comprehensive correlation study.
Preparation of 3D micro-tumors from PDTX mice and correlation to treatment efficacy in PDTX mice:
Colorectal xenograft samples were obtained from PDTXs previously established at the University of Colorado Cancer Center, Denver, CO, USA. Briefly, fresh colorectal tumor tissue was obtained from consenting patients at the University of Colorado Hospital in accordance with protocols approved by the Colorado Multiple Institutional Review Board. Tumor samples were cut into 3 mm3 pieces and implanted into the flank of 4-6 weeks old female athymic nuce mice. When tumor volume reached 1000-1500 mm3, the tumor was excised and passaged into subsequent generations. For micro-tumor preparation, tumors from untreated mice were excised, placed in cold PBS with antibiotics and transported to the laboratory on ice. All animal experiments were performed under an approved protocol by the Institutional Animal Care and Use Committee.
We obtained a 86% success rate (12 out of 14 tested) in generating micro-tumors from PDTX mice.
Three animals were treated and showed almost complete response (70-80% remission) when treated with Irinotecan. Tissue from tumors of these animals were tested for treatment sensitivity in the ChemoGuide test (Fig. 4.3.1).
A low number of animals tested did not allow a correlation analysis to be made; but it could be concluded that with the ChemoGuide technology it is possible to produce micro-tumors from the PDTX mice, and that it was observed that patient micro-tumors were sensitive to Irinotecan as observed in the treated mice.
WP 5: Evaluation of the correlation between test results and patient out-come
Short description of planned work from DoW:
As part of the standard protocol for patients receiving adjuvant chemotherapy treatment for colorectal cancer the treatment efficacy will be evaluated with CT-scans three and six months after initiation of chemotherapy.
The RECIST response criteria will be used to evaluate the disease response or progression following the chemotherapy. The result of a ChemoGuide screening is a predication of tumor resistance and sensitivity towards chemotherapy for the individual patient. By comparing the individual patient's ChemoGuide screening prediction and the RECIST classification as monitored by CT-scanning, it will be possible to verify the prognostic value of the ChemoGuide screening system. Importantly significant correlation will demonstrate that the ChemoGuide screening system is highly useful for prospective chemo-sensitivity screening.
At the completion of the patient screening (WP4) the treatment efficacy data from each individual patient will be collected, this includes CT-scans, the RECIST evaluation and other medical evaluations performed by the oncologists.
Once the patient data has been collected, a correlation between the individual patient’s treatment efficacy and the ChemoGuide screening results will be performed. ChemoGuide predictions of both sensitivity and resistance towards the treatment regimens will be used in this correlation.
Summary of WP results:
In total we enrolled 253 patients in the ChemoGuide project. Of these 102 patients where included in the clinical study. The screening results have been analyzed using image analysis where the growth of the 3D micro-tumors have been measured and categorized in three sensitivity groups: No response (resistance); Partial response; Complete response.
The full correlation between screening results and patient outcome will be conducted when all stage IV and stage IIB/III patients have been evaluated using the RECIST and the Disease-Free-Survival (DFS) criteria, respectively.
In the original DoW we anticipated that at least 50 patients would be enrolled in Denmark, Germany and the US. This has, as presented in the approved amendment, been changed such that all clinical studies involving human patients will be conducted in Denmark and Germany. This change has not influenced the total number of patients enrolled in the project. In the laboratory of our US-partner we have instead expanded the ChemoGuide project by including experiments with Patient-Derived-Tumor-Xenografs in mice. These new experiments have resulted in addition of a new deliverable: 4.3 Correlation between screening data and treatment outcome in PDTX.
In DoW we presented that the screening results would be correlated to the RECIST criteria monitored using CT scanning 3 and 6 months after launch of adjuvant treatment. This will still be the case for stage IV patients with liver metastasis however; it cannot be done in the stage IIb/III patients as all patients we received had got their primary tumor completely resected. For this patient group we will correlate the screening results with Disease-Free-Survival (DSF). As this patient group has an overall survival is significantly better than the stage IV patients (see Fig. 5.1.1) we will need to wait to conduct the final correlation between test results and patient outcome until two years after launch of the adjuvant treatment for the last enrolled patient. For stage IV patients this correlation can be conducted two years earlier.
Deliverable 5.1: Screening data evaluated and correlation to patient outcome concluded:
Screening process and data acquisition:
The clinical screening in the ChemoGuide project is conducted using a 3D image scanning reader (oCelloscope). As mentioned in RP1 we decided early in the ChemoGuide project to move from a 2D cell format to 3D. This decision led us to investigate a new image-based reader system that allows fast detection of 3D information in bright field illumination. Using a novel scanning technology, the reader is capable of collecting the 3D information in a single scan.
The screening process is shown schematically in Fig. 5.1.2.
After setting up the 3D micro-tumors in the arrays we follow the growth for seven days with daily scans. In the final product we may lower the number of scans as the amount of data collected is very high (approx. 35 Gb per patient).
Important for this test is that the 3D micro-tumors are perturbed as little as possible. We have therefore omitted use of molecular probes for monitoring metabolic functions, and conducted an end-point measure of growth inhibition evoked by the different treatment regimens.
The image analysis is conducted using a semi-automatic image analysis algorithm developed in collaboration with Philips BioCell who as developed the 3D reader (oCelloscope, http://www.philips.com/content/corporate/en_AA/biocell/home.html).
Data analysis:
Images acquired by the oCelloScope are analyzed in a two-step process. Frist an automated algorithm stiches the images together and generates z-stacks, it then identifies 3D micro-tumors using edge finding, contrast based methods while also incorporating the 3D information in the acquired images.
A secondary program is used to manually check the results of the automated algorithm and is used to remove false positives and add 3D micro-tumors that were not correctly identified using alternative micro-tumor-finding algorithms. When all micro-tumors are identified their 2D areas are calculated. (Fig. 5.1.3).
The area is monitored daily over seven days (Fig. 5.1.4).
From Fig. 5.1.4 it can be seen that the growth rate varies from patient to patient. This is a reflection of the patient heterogeneity that not only is observed with regard to the sensitivity/resistance to certain treatment regimens, it is also observed when monitoring the growth potential of the patient’s tumor. It is therefore very important that each patient is its own control when evaluating the effectiveness of the different treatment regimens. This is exemplified in Fig. 5.1.5 where the responsiveness of three patients towards Folfox and Folfiri (with and without Cetuximab) is monitored.
From Fig. 5.1.5 it is observed that Pt601 show low sensitivity to all four treatment regimens, whereas Pt606 show moderate sensitivity to Folfox with no extra benefit of including Cetuximab. The last Pt574 is a highly sensitive patient where efficacy is observed for all four treatments with noticeable improvement of including Cetuximab.
Using the data presentation shown in Fig. 5.1.4 the responsiveness of the individual patient to each treatment regimen will be grouped in: No response (resistance); Partial response; Complete response.
This grouping will be correlated to the RECIST (stage IV patients) and the Disease-free-survival (stage IIb/III patients) (see below).
Correlation of screening results with patient outcome:
In total we have enrolled 253 patients in the ChemoGuide project. The enrolment started in June 2014. Of these patients, we successfully prepared 3D micro-tumors from 102 patients and included these in the screening activity. The reason for the relatively large number of patients that could not be used for screening is explained in Deliverable 4.1. On average we have been able to run two arrays on each patient (in total 254 arrays). All most half (46%) of the included patients had metastatic disease (stage IV with liver metastasis) (Deliverable 4.1 Table 4.1.1).
The results of the clinical screening conducted in the ChemoGuide project will not at this early stage influence which treatment regimen is offered to the patients. The purpose of the study is to show whether there is a correlation between the screening results and the responsiveness of the patient to the given treatment. The screening results will be associated with clinical staging and associated biomedical data of the individual patient. The screening results for the patients with metastatic disease (stage IV with liver metastasis) will be correlated to the RECIST evaluation. The RECIST measurements will be based on CT scans performed prior to the initiation of chemotherapy and 3 and 6 months after the initiation of chemotherapy. In short, the RECIST evaluation will measure lesion diameters and their development, the result of the RECIST evaluation will be a division of patients into four categories depending on their response to the chemotherapy treatment:
• Complete Response (CR): Disappearance of all target lesions. Any pathological lymph nodes must have reduction in short axis to <10 mm.
• Partial Response (PR): At least a 30% decrease in the sum of diameters of target lesions, taking as reference the baseline sum diameters.
• Stable Disease (SD): Neither sufficient shrinkage to qualify for PR nor sufficient increase to qualify for PD, taking as reference the smallest sum diameters since the treatment started
• Progressive Disease (PD): At least a 20% increase in the sum of diameters of target lesions, taking as reference the smallest sum on study (this includes the baseline sum if that is the smallest on study). In addition to the relative increase of 20%, the sum must also demonstrate an absolute increase of at least 5 mm.
For the stage IIB/III patients (54%) the RECIST criteria cannot be used when evaluating the treatment efficiency as all these patients have got their primary tumor completely resected. The clinical endpoint to be used for these patients will be Disease-free-survival monitored two years after launch of the adjuvant treatment.
For both patient categories we are not allowed (due to the blinded design of the study) to conduct the correlation between screening results and patient outcome before all stage IV patients have been through the 6 month CT scanning and the stage IIb/III patients have been followed for two years after start of treatment.
Example of how to correlate screening results to RECIST evaluation is presented in Fig. 5.1.6.
WP 6: Dissemination
Short description of planned work from DoW:
The oncologists of the ChemoGuide project will present the test findings for colleagues at their own clinical center and at corresponding center in their respective countries. Further it is planned to present the findings at international meetings organized by e.g. the European Cancer Organisation and the American Society of Clinical Oncology.
The commercialization of the ChemoGuide test will have to be supported by strong clinical evidence. The best way scientifically to present that is through peer reviewed international journals.
The clinical partners of the present project represent the end-users of the final test. The commercialization of the ChemoGuide test will therefore initially occur at these test centers and similar sites in Europe and USA. To support this commercialization, we will prepare descriptive material that presents the scientific/clinical concept behind the test and how the test results are going to support the tailoring of individualized adjuvant treatment.
This material will include semi-scientific reports, flyers and a web-site.
Summary of WP results:
The technology and the clinical findings (see Deliverable 5.1) have been presented to oncologists at the two partnering hospitals (Bispebjerg Hospital and Universitaetsklinikum Hamburg-Eppendorf) and other hospitals in the Copenhagen and Hamburg region. This has increased the number of hospitals involved from the present two to seven. Even though a complete correlation analysis between screening results and patient outcome is not yet possible (see Deliverable 5.1) the tremendous interest clearly shows a belief in the technology and clinical unmet need.
Discussions are ongoing with further oncology centers in the participating countries and in Sweden.
Deliverable 6.1: Presentation of clinical finding
Presentation of clinical findings:
The presentation and subsequent commercialization of the ChemoGuide screening test will occur through the established test centers in Denmark and Germany. To increase the knowledge of this new screening technology in the clinical environment we have several meetings with oncology colleagues at the partnering hospitals and beyond. The information presented has been very positively received and resulted in continue collaboration between the ChemoGuide partners and inclusion of additional hospitals in both Germany and Denmark (Hvidovre Hospital and Hillerød Hospital) that will recruit further patients for the ChemoGuide analysis.
Technical and clinical findings have been discussed among the ChemoGuide project members in numerous Skype meetings and several face-to-face meetings (2. April 2014; 16. April 2014; 12-13. May 2014; 28-31. July 2014; 22-23. Oct. 2014; 19. Dec 2014; 14. Jan 2015; 9-10. Febr 2015; 1. April 2015; 15-16. April 2015; 9. June 2015). This very active interaction between the partners has been key to the success of this multidisciplinary project.
In the ChemoGuide project we have prioritized presenting the technology to clinical and pharmaceutical institutions to support a future commercialization rather than writing scientific publications. The resemblance between 3D micro-tumors developed from liver metastasis and the lesion in-vivo has however, been presented at the American Association of Cancer Research 2014 meeting (San Diego, 4-11 April 2014; see Fig. 6.1.1).
Deliverable 6.2: Commercial product description
The partners behind the ChemoGuide project will continue (after finalization of the EU-funded project) the collaboration towards commercialization of the ChemoGuide test. Laboratory facilities at Bispebjerg Hospital (Copenhagen) and at Universitaetsklinikum Hamburg-Eppendorf (Hamburg) will thus continue to conduct follow-up evaluation of the enrolled patients to evaluate the predictive power of the test. MicroLiquid (Bilbao) will continue the production of the latest test arrays.
Commercial promotion of the test has been conducted latest at the BIO conference in June 2015 Philadelphia (USA). A semi-scientific description of the technical/clinical concept behind the test has been prepared through a Teaser and a Slide deck (see Fig. 6.2.1).
The web-site for 2cureX has been updated to include data from the ChemoGuide project.
Potential Impact:
The impact of the ChemoGuide project is separated into the following five topics:
1. 3D tumoroid based functional testing show value in predicting patient outcome after specific chemotherapy treatment
2. The ChemoGuide technology is applicable to other cancers than colorectal cancer
3. Dissemination of results
4. Regulatory requirements defined
5. Basis for commercial benefit to the involved SMEs
6. Basis for commercial benefit to the European society
Ad 1: 3D tumoroid based functional testing show value in predicting patient outcome after specific chemotherapy treatment.
In total we enrolled 253 patients in the ChemoGuide project. Of these 102 patients where included in the clinical study. The screening results have been analyzed using image analysis where the growth of the 3D micro-tumors have been measured and categorized in three sensitivity groups: No response (resistance); Partial response; Complete response.
The prospective, non-interventional clinical study conducted in the ChemoGuide project is designed such that all used treatment regimens are included in the screening array; but the result of the ChemoGuide test does not influence the selection of treatment. The patients received standard treatment as proposed by the oncologist and when patient outcome is available it will be investigated whether the ChemoGuide test correctly predicted outcome. Preliminary observations clearly suggest that such positive correlation exists (see Deliverable 5.1). Even though full analysis of the predictiveness of the test cannot be revealed earlier than 2-3 years after enrolment of the last patient in the study, oncologists from the partnering cancer centers designed an interventional study on stage IV metastatic colorectal cancer patients where the metastasis (most often liver) cannot be resected. This group constitutes the largest proportion of the stage IV patients, and the patients having the poorest prognosis.
The proposed study will be conducted at the partnering hospitals in Denmark and Germany plus one of the largest cancer centers in the United Kingdom. It is a major endorsement of the ChemoGuide test that the oncologists this early will embark on using the test in designing the treatment regimen for the individual patient.
The impact for the patients will be significant as the stage IV colorectal cancer patients have very little time left for selecting an efficacious treatment. Colorectal cancer is one of the most diagnosed cancers in Europe. Alone in the three countries where the intervention study will be conducted more than 20.000 colorectal cancer stage IV patients are diagnosed yearly.
Not only will a fast selection of an efficacious treatment benefit the patient, it will also result in major cost savings for the health care systems.
In the original DoW for the ChemoGuide project we described how genomic profiling had become synonymous with Personalized medicine: “The concept of individualized treatment – personalised medicine – was introduced as a conceptual part of the genome project more than two decades ago. The ability to design an individualized treatment therefore became synonymous with correlating the patient’s genetic profile to his/her’s drug responsiveness. Genotyping of patients has proven successful in identification of responsive patients to single targeted drugs like Herceptin (HER-2/neu expression) and Erbitux/Tarceva (KRAS or BRAF mutation). However, this has not been the case when trying to match specific combination therapies to individual patients.”
Since the launch of the ChemoGuide project the notion that genomic information is insufficient in tailoring the treatment of the individual patient has received significant support (for recent publications see Le Tourneau et al. (2015) Molecularly targeted therapy based on tumor molecular profiling versus conventional therapy for advanced cancer (SHIVA): a multicentre, open-label, proof-of-concept, randomised, controlled phase 2 trial. The Lancet 16, 1324-34; Friedman et al. (2015) Precision medicine for cancer with next-generation functional diagnostics. Nature Rev Cancer 15, 747-756).
The need and impact for clinically validated functional test will increase dramatically in the coming years. Here we see the test developed and clinically tested in the ChemoGuide to be in the forefront of cell-functional tests.
Ad 2: The ChemoGuide technology is applicable to other cancers than colorectal cancer
In the ChemoGuide project we have concentrated our effort on colorectal cancer. However, a subsequent introduction of the technology to other types of cancer is important both a commercial and societal point of view.
During the ChemoGuide project other cancers like pancreas, liver, breast and head & neck were tested for our ability to prepare 3D micro-tumors. It all cases it turned out positive. In conclusion it is likely that the preparation of 3D micro-tumors developed in ChemoGuide is applicable to other solid tumors.
It is obvious that expanding the test into other solid cancer will greatly increase the impact of the ChemoGuide test.
Ad 3: Dissemination of results
The ChemoGuide project has a clear commercial objective of providing a test for tailoring the medical treatment of colorectal cancer patients. It has therefore been the focus of the project group to protect developments and finding by patent filings. Subsequently the project will pursue dissemination through scientific publications.
One patent was filed in 2014 by 2cureX. Hagel and Thastrup: Identifying compounds modifying a cellular phenotype. PCT/DK2015/050197. The discoveries behind the patent is solely made by 2cureX and related to the integration of technologies for testing of patient material and for patient diagnostics. The patent is within the competences and business strategy of 2cureX. 2cureX is therefore the sole owner of the patent in accordance with the Consortium Agreement:
“8.1 Foreground ownership
RTD Provider Foreground (e.g. Intellectual Property) developed during the course and falling within the Field shall be vested to and owned by one or more of the SMEs. Transfer of rights in RTD provider Foreground to one or more of the SMEs shall be granted on Fair and Reasonable conditions. RTD Provider Foreground which is Patient Material shall at all times vest in RTD Provider and shall principally not be transferable, unless specified by a separate Agreement. The distribution of the RTD Provider Foreground between the SMEs is to be decided by Executive Board, along the protocol described below. It will be the responsibility, including the financial obligation, of the relevant SME(s) to file and prosecute associated patent applications. The relevant SME(s) will pay close attention to who the actual inventors are. The RTD and the associated inventors will be responsible for delivering all material necessary for the patent filing.
Foreground ownership shall be determined along the following principles, or as, exceptionally, otherwise agreed by the Executive Board:
2CX shall receive Foreground that is directly related to the integration of technologies for testing of patient material and for patient diagnostics and 2CX shall receive Foreground related to testing of patient material;
MLD shall receive Foreground that is directly related to the manufacturing of polymer devices and associated injection moulding procedures.
In case of conflicts with regard to the distribution of the ownership the Executive Board will decide.
The relevant SME has the obligation including bearing the associated cost of protecting the Foreground. In case no SME wishes to comply with this obligation the ownership shall remain with the relevant RTD provider.”
Ad 4: Regulatory requirements defined
The regulatory requirements needed for commercialization of the ChemoGuide test is the following:
• Establishment of an EN ISO 13485 compliant quality management system (QMS)
• Obtaining the certification of the EN ISO 13485 QMS by a relevant certification body (TÜV Süd)
• Establishment of the required technical documentation for the ChemoGuide test
• Ensure concordance with the IVD directive 98/79/EC
• Presentation of the clinical validation results to demonstrate the benefit of ChemoGuide test for the patient and the health care system
• Issuing a declaration of conformity for the ChemoGuide test
2cureX has identified the company, Simply Quality – Qualitätsmanagementberatung in Weilheim, Germany to manage the process of getting above certifications obtained.
Ad 5: Basis for commercial benefit to the involved SMEs
Two SMEs (2cureX and MLD) we partners in the ChemoGuide project. MLD has been responsible for optimization, prototyping and mass production of the disposable arrays used in the ChemoGuide test whereas 2cureX has assembled the ChemoGuide arrays with the treatment regimens and distributed them to the clinical partners for testing.
The successful production of the disposable arrays by MLD opens for commercial benefit to the SME by being the future mass producer of the arrays. Expansion into other types of cancers will require production of a large number of arrays. Further the knowledge of producing arrays for 3D cell cultures will become valuable as research tool in other biological and therapeutic areas.
2cureX is planning a subsequent interventional study with clinical partners in Denmark, Germany and the United Kingdom. The ChemoGuide project has paved the way for conduction of this final clinical study before commercialization into the clinical environment. Important for the subsequent roll-out of the test in major European cancer clinics are that the already involved partners are key opinion leaders in each their country.
It is essential for the commercial success of the test that the clinical partners of the ChemoGuide project has agreed to present the predictive power of the test to other major oncology centers in Europe.
2cureX will provide the integrated test with disposable arrays containing treatment regimens and test procedure. Data analysis can be conducted on-site or 2cureX can provide the service and as part thereof compare the results of the individual test with data from a steadily expanding database.
List of Websites:
www.chemoguide.eu