Periodic Report Summary 1 - MELENDO2 (Genetic epidemiological investigation into the association between endometriosis and cutaneous melanoma)
1.1 Project objectives
The objective of this project is to investigate the potential genetic overlap between endometriosis and cutaneous melanoma, with 5 specific aims:
Aim 1: To explore the association between endometriosis and melanoma
Aim 2: To perform a pathway-based analysis in genome-wide association studies (GWAS) of melanoma to identify the biological pathways that influence melanoma risk
Aim 3: To test melanoma-associated pathways/SNPs in relation to endometriosis
Aim 4: To test endometriosis pathways/SNPs in relation to melanoma in an American population
Aim 5: To test endometriosis pathways/SNPs in relation to melanoma in a French population
1.2 Training
The focus of the outgoing phase of the project was on training in genetic epidemiology, and in programming in SAS under Unix using the Channing system and macros. The fellow audited 4 courses in genetic epidemiology at Harvard School of Public Health during the 2012-2013 academic year. These courses provided her with extended and up-to-date knowledge in statistical methodologies applied to genetic epidemiological research, including in the pathway-based analysis technique, for which she received practical training during the second year of the project. The fellow was also trained to use the complex programming macros developed at the Channing to develop her skills in time-varying proportional hazards modeling. After attending theoretical lectures and webinars, the fellow has been practicing her new skills by working on 2 side projects relevant to the project aims, exploring respectively 1) pigmentation and family history of melanoma, and 2) UV exposure in relation to endometriosis risk in the NHSII cohort. To complete her methodological training, the fellow also attended numerous scientific seminars in her research field, and she also sought and attended many career development workshops.
1.3 Progress and main results
Aim 1
Following the training phase of the project, the fellow has been exploring the associations between endometriosis and cutaneous melanoma in the NHSII cohort and initially produced preliminary results (a significantly positive association between endometriosis and melanoma was observed in age-adjusted models (relative risk (RR)=1.38 95% confidence intervals (CI)=1.02-1.88) although it was slightly attenuated after adjustment for skin cancer risk factors (RR=1.28 95% CI=0.94-1.74)). However, in 2014, the Endometriosis Group of the Channing has re-thought its approach to code endometriosis diagnosis within the Harvard cohorts. The fellow co-led this programming effort to refine endometriosis case definition and documented the programming decisions, providing guidelines and methodological instructions for the Group. This step resulted in an additional ~500 endometriosis cases to be considered in the analysis of Aim 1; because of this major change, the data needed to be re-analyzed in order to explore the association between endometriosis and cutaneous melanoma more accurately. The analysis and corresponding manuscript are currently underway and will be submitted to an international peer-reviewed journal in 2015. Relevant to this aim, the fellow performed the analyses for 2 side projects to explore the relationships between host factors, UV exposure and endometriosis risk in the NHSII cohort. The first project showed increased endometriosis risk associated with presence of nevi, higher level of skin’s burning reaction to sun exposure and family history of melanoma, which further suggest a potential common genetic background between endometriosis and melanoma (Kvaskoff et al., Int J Epidemiol 2014). The second project showed a higher endometriosis risk associated with measures of intense recreational sun exposures, but a decreased risk associated with residential sun exposure. Because of the increase in endometriosis case numbers, these analyses also needed to be updated; the fellow is currently completing this step and performing sensitivity analyses to explore the potential confounding effect of vitamin D on these associations. The manuscript successfully passed internal Program and Technical reviews and is currently being finalized.
Aim 2
The fellow audited a course on the pathway-based analysis technique at Harvard School of Public Health and has been in touch with the investigators who have expertise in this method. She then received practical training on this technique by Xin Li in Dr. Han’s group whose work centered around a pathway-based analysis for basal-cell carcinoma. The fellow will be in close contact with Ms. Li as she performs the analysis for melanoma, which will be performed in 2015.
Aim 3
The fellow performed a comprehensive literature review of the SNPs associated with melanoma, pigmentation and nevus propensity, which formed the basis for the selection of the SNPs to be genotyped. The genotyping phase is complete and the data are ready to be analyzed. For this aim, the fellow and Dr. Missmer initiated a further collaboration with Prof. Grant Montgomery from QIMR Berghofer Medical Research Institute in Australia. The fellow will investigate this topic in a combined analysis with data from the NHSII cohort, the endometriosis GWAS from QIMR, and the endometriosis GWAS from the OXEGENE study at Oxford (UK). The fellow will perform the analysis in 2015, and the results will be combined in a single, higher-impact manuscript.
Aim 4
The fellow performed a comprehensive literature review of the SNPs associated with endometriosis in the available endometriosis GWAS studies. To increase the statistical power of the analyses, the fellow will test the associations in the Australian melanoma GWAS data available from QIMR, in addition to the American melanoma GWAS data in the NHS cohort. These analyses will be jointly performed with those related to Aim 3 and will be combined in a single manuscript (see above).
Aim 5
Genetic variants were selected based on the fellow’s comprehensive literature search (see Aim 3). The selection of SNPs, genotyping, and all quality control checks were performed and are complete. Following the hypotheses developed in Aim 4, an additional set of SNPs was genotyped in the French population in order to further test the associations between endometriosis-related SNPs and melanoma risk in E3N. These analyses will be performed during the return phase of the project.
1.4 Expected final results and impact
Previous studies strongly suggest that endometriosis patients are at higher melanoma risk, but the causes underlying this relationship are currently unknown. There is increasing evidence suggesting a common genetic basis for the two diseases, but no previous research has investigated this intriguing aspect. However, it is essential to identify which genes and pathways are involved in both conditions in order to understand what is common to endometriosis and melanoma patients, and to potentially discover further treatment options for one or both diseases.
The expected final results from this project are 1) to confirm an association between endometriosis and melanoma; 2) to discover biological pathways associated with melanoma risk; and 3) to find specific associations between endometriosis-associated genetic variants and melanoma risk, and between melanoma-associated variants and endometriosis risk.
In the long-term, this project will contribute to enhance our knowledge of the genetic background of melanoma and endometriosis and the factors associated with their development. This knowledge will inform therapies targeted at the patients developing both diseases, but also each of them separately, and may be used in personalized medicine in the future. Overall, the findings from this project will uniquely contribute to increase our understanding of the association between endometriosis and melanoma, two major health burdens that affect millions of women in Europe and beyond.
The objective of this project is to investigate the potential genetic overlap between endometriosis and cutaneous melanoma, with 5 specific aims:
Aim 1: To explore the association between endometriosis and melanoma
Aim 2: To perform a pathway-based analysis in genome-wide association studies (GWAS) of melanoma to identify the biological pathways that influence melanoma risk
Aim 3: To test melanoma-associated pathways/SNPs in relation to endometriosis
Aim 4: To test endometriosis pathways/SNPs in relation to melanoma in an American population
Aim 5: To test endometriosis pathways/SNPs in relation to melanoma in a French population
1.2 Training
The focus of the outgoing phase of the project was on training in genetic epidemiology, and in programming in SAS under Unix using the Channing system and macros. The fellow audited 4 courses in genetic epidemiology at Harvard School of Public Health during the 2012-2013 academic year. These courses provided her with extended and up-to-date knowledge in statistical methodologies applied to genetic epidemiological research, including in the pathway-based analysis technique, for which she received practical training during the second year of the project. The fellow was also trained to use the complex programming macros developed at the Channing to develop her skills in time-varying proportional hazards modeling. After attending theoretical lectures and webinars, the fellow has been practicing her new skills by working on 2 side projects relevant to the project aims, exploring respectively 1) pigmentation and family history of melanoma, and 2) UV exposure in relation to endometriosis risk in the NHSII cohort. To complete her methodological training, the fellow also attended numerous scientific seminars in her research field, and she also sought and attended many career development workshops.
1.3 Progress and main results
Aim 1
Following the training phase of the project, the fellow has been exploring the associations between endometriosis and cutaneous melanoma in the NHSII cohort and initially produced preliminary results (a significantly positive association between endometriosis and melanoma was observed in age-adjusted models (relative risk (RR)=1.38 95% confidence intervals (CI)=1.02-1.88) although it was slightly attenuated after adjustment for skin cancer risk factors (RR=1.28 95% CI=0.94-1.74)). However, in 2014, the Endometriosis Group of the Channing has re-thought its approach to code endometriosis diagnosis within the Harvard cohorts. The fellow co-led this programming effort to refine endometriosis case definition and documented the programming decisions, providing guidelines and methodological instructions for the Group. This step resulted in an additional ~500 endometriosis cases to be considered in the analysis of Aim 1; because of this major change, the data needed to be re-analyzed in order to explore the association between endometriosis and cutaneous melanoma more accurately. The analysis and corresponding manuscript are currently underway and will be submitted to an international peer-reviewed journal in 2015. Relevant to this aim, the fellow performed the analyses for 2 side projects to explore the relationships between host factors, UV exposure and endometriosis risk in the NHSII cohort. The first project showed increased endometriosis risk associated with presence of nevi, higher level of skin’s burning reaction to sun exposure and family history of melanoma, which further suggest a potential common genetic background between endometriosis and melanoma (Kvaskoff et al., Int J Epidemiol 2014). The second project showed a higher endometriosis risk associated with measures of intense recreational sun exposures, but a decreased risk associated with residential sun exposure. Because of the increase in endometriosis case numbers, these analyses also needed to be updated; the fellow is currently completing this step and performing sensitivity analyses to explore the potential confounding effect of vitamin D on these associations. The manuscript successfully passed internal Program and Technical reviews and is currently being finalized.
Aim 2
The fellow audited a course on the pathway-based analysis technique at Harvard School of Public Health and has been in touch with the investigators who have expertise in this method. She then received practical training on this technique by Xin Li in Dr. Han’s group whose work centered around a pathway-based analysis for basal-cell carcinoma. The fellow will be in close contact with Ms. Li as she performs the analysis for melanoma, which will be performed in 2015.
Aim 3
The fellow performed a comprehensive literature review of the SNPs associated with melanoma, pigmentation and nevus propensity, which formed the basis for the selection of the SNPs to be genotyped. The genotyping phase is complete and the data are ready to be analyzed. For this aim, the fellow and Dr. Missmer initiated a further collaboration with Prof. Grant Montgomery from QIMR Berghofer Medical Research Institute in Australia. The fellow will investigate this topic in a combined analysis with data from the NHSII cohort, the endometriosis GWAS from QIMR, and the endometriosis GWAS from the OXEGENE study at Oxford (UK). The fellow will perform the analysis in 2015, and the results will be combined in a single, higher-impact manuscript.
Aim 4
The fellow performed a comprehensive literature review of the SNPs associated with endometriosis in the available endometriosis GWAS studies. To increase the statistical power of the analyses, the fellow will test the associations in the Australian melanoma GWAS data available from QIMR, in addition to the American melanoma GWAS data in the NHS cohort. These analyses will be jointly performed with those related to Aim 3 and will be combined in a single manuscript (see above).
Aim 5
Genetic variants were selected based on the fellow’s comprehensive literature search (see Aim 3). The selection of SNPs, genotyping, and all quality control checks were performed and are complete. Following the hypotheses developed in Aim 4, an additional set of SNPs was genotyped in the French population in order to further test the associations between endometriosis-related SNPs and melanoma risk in E3N. These analyses will be performed during the return phase of the project.
1.4 Expected final results and impact
Previous studies strongly suggest that endometriosis patients are at higher melanoma risk, but the causes underlying this relationship are currently unknown. There is increasing evidence suggesting a common genetic basis for the two diseases, but no previous research has investigated this intriguing aspect. However, it is essential to identify which genes and pathways are involved in both conditions in order to understand what is common to endometriosis and melanoma patients, and to potentially discover further treatment options for one or both diseases.
The expected final results from this project are 1) to confirm an association between endometriosis and melanoma; 2) to discover biological pathways associated with melanoma risk; and 3) to find specific associations between endometriosis-associated genetic variants and melanoma risk, and between melanoma-associated variants and endometriosis risk.
In the long-term, this project will contribute to enhance our knowledge of the genetic background of melanoma and endometriosis and the factors associated with their development. This knowledge will inform therapies targeted at the patients developing both diseases, but also each of them separately, and may be used in personalized medicine in the future. Overall, the findings from this project will uniquely contribute to increase our understanding of the association between endometriosis and melanoma, two major health burdens that affect millions of women in Europe and beyond.