Multiple primary melanoma in association with other personal and family history of cancers.

Abstract

e21559 Background: Patients (PTs) with cutaneous melanoma are at increased risk of developing second primary melanoma and non-melanoma skin cancers. The primary aim of this study was to define the association between MPM and personal history of non-melanoma skin cancers and other non-skin cancers. The secondary aim was to evaluate the association between MPM and the presence of other cancers among first-degree relatives (FDRs). Methods: We performed a retrospective case-control study including cases with MPM and controls with single primary melanoma (SPM) from the University of Pittsburgh Cancer Institute Melanoma Center Biological Sample and Nevus Bank. The proportions and percentages of non-melanoma skin cancer, other non-skin cancer, 1st degree family history of melanoma, and 1st degree family history of other non-melanoma cancers were calculated separately for MPM and SPM groups. Fisher’s exact tests were performed to test whether MPM was associated with these variables. For each significant variable, a multivariable logistic regression model was used to test its association with MPM after adjusting for age, gender, melanoma staging, and smoking status. Results: In total, 311 PTs (39.2% men; median age at initial diagnosis 51years) were enrolled, including 194 with SPM (38.6%; 51) and 117 with MPM (39.8%; 48). 28 (9%) of PTs had squamous cell carcinoma (SCC), and 63 (20%) had basal cell carcinoma (BCC). The most common non-skin cancers in the whole cohort were prostate (4.8%), breast (3.8%), hematological (1.9%), colorectal (1.3%), and cervical cancers (1.3%). FDR history of melanoma, non-melanoma skin cancer, and other cancers were positive in 15.4%, 7.1% and 46.3% PTs, respectively. The most common non-skin cancers in FDRs were breast, prostate, lung, colorectal and hematological malignancies. In comparison to PTs with SPM, PTs with MPM were more likely to have SCC (14.5% vs 5.7%, p=0.013) but not BCC and other non-skin cancers. FDRs of PTs with MPM had higher prevalence of melanoma (23.1% vs 10.8%, p=0.005), prostate cancer (31.9% vs 5.3%, p=0.0002) but not other non-melanoma skin and non-skin cancers. In multivariate analysis the association remained significant between MPM and SCC (OR 2.7, 95% CI 1.1-6.6, p=0.032), FDR history of melanoma (OR 2.0, 95% CI 1.03-4.1, p=0.042), and FDR history of prostate cancer (OR 5.6, 95% CI 1.6-20.3, p=0.008). Conclusions: MPM is associated with higher prevalence of SCC and FDR history of melanoma and prostate cancer, but not BCC and other non-melanoma cancers in comparison to SPM.