Second primary malignancies in multiple myeloma: Incidence, survival and prognostic factors from a large single-center observational study

Abstract

Abstract Introduction: The risk of second primary malignancies (SPMs) has emerged as a growing clinical concern in patients with multiple myeloma (MM) who currently achieve prolonged survival. The reported cumulative incidence of SPMs in MM has varied greatly depending on patient characteristics, treatment and stringency of follow up. we sought to estimate the incidence of SPM in a large MM population followed long-term in a single center and identify risk factors for SPM, including genetic aberrations. Methods: This retrospective cohort study includes patients (pts) with MM who initiated first-line Rx at Memorial Sloan Kettering Cancer Center, excluding patients who did not continue care at MSK beyond induction and transplant, and those internally referred to avoid selection bias inherent to a cancer center. The primary objective is to estimate the cumulative incidence of SPM following Rx start, excluding non-invasive skin cancers or recurrence of prior known malignancies, and explore risk factors for SPM. Secondary objectives are: (1) To evaluate associations between MM and specific SPMs; (2) To analyze overall survival (OS) of pts with SPM. The Kaplan-Meier method estimated median OS. Cumulative incidence of SPM was calculated with death as a competing risk. Cause-specific Cox models estimated hazard ratios for SPM risk, accounting for competing risks, and landmark analysis assessed the association between SPM development and OS. Results: We included 1,476 patients with a median follow-up of 5.7 years. The median OS for all pts is 15.6 years (95% CI, 14.1–Not reached); (5-year OS: 83.8% (95% CI, 81.6%-86%) and 10-year OS: 65.5% (95% CI, 61.6%-69.7%). Median age at start of Rx is 64 years (IQR, 57–71). 54% are male. Racial distribution includes 72% Caucasian, 15% African American, 5% Asian, and 8% others. A smoking history is present in 37%, history of other malignancy prior to MM diagnosis in 18%, and a first-degree family history of cancer in 56%. The isotype is IgG (61%), IgA (21%), and IgD (1%), and light chain only (17%) with ISS stages I, II, and III in 48%, 34%, and 18%. High-risk cytogenetic features at diagnosis include gain/amp 1q (28%), del17p (10%), t(4;14) (7%), and t(14;16) (2%). 238 patients acquired additional cytogenetic abnormalities during follow up. Induction Rx included VRD (29%), KRD (20%), DVRD (16%), DKRD (8.1%), VCD (4.2%), and others (22.7%). ASCT occurred in 55%, and CAR-T cell Rx in 7%. The 5- and 10-year estimated cumulative incidence of all-SPMs is 8.8% (95% CI, 7.3–11%) and 20% (95% CI, 17–23%), respectively (Heme-SPMs in 1.7% and 5.8%, and solid-SPMs in 7.1% and 13%, 5 and 10 years respectively). The median number of lines of therapy prior to SPM is 1 (IQR 1–2). In those patients with SPM, the median time to SPMs after start of Rx is 3.7 years (IQR 1.8–6.7): 5.5 years (IQR 3–7.8) in Heme-SPM and 3 years (IQR 1.5-6.2) in Solid-SPM. Heme-SMPs (n=42) include MDS 42%, ALL 28%, AML 11%, NHL 9%, MPD 7%, and LGL 2%. Solid-SPMs (n=129) include prostate 17%, GU 14%, lung 14%, melanoma 12%, breast 9%, GI 9% and other cancers 24%. Although univariate analysis showed a trend toward increase in all-SPM risk for patients with prior malignancy (HR 1.43, 95% CI 0.99–2.1) and male sex (HR 1.3, 95% CI 0.98–1.8); and association of heme-SPM risk with ASCT (HR 2.0, 95% CI 0.99–3.8), multivariable model, adjusted for significant univariable factors, identified age 55–65 (HR 1.7, 95% CI 1.1–2.8), age ≥65 (HR 2.1, 95% CI 1.3–3.4), Caucasian race (HR 1.6, 95% CI 1.0–2.7), and smoking history (HR 4.7, 95% CI 1.7–13) as independent predictors of SPM. The median OS after SPM diagnosis was 3.0 years (95% CI 2.3–NA) for Heme-SPMs and 12.1 years (95% CI 6–NA) for solid-SPMs. Conclusions: Acknowledging the limitations inherent to a tertiary referral cancer center and related biases, this study shows in pts with MM, the 5 and 10-year estimated cumulative incidence of SPM after start of MM Rx, is 8.8% and 20%, respectively. Notably, 18% of patient also have prior malignancies diagnosed prior to MM. Patients who developed solid SPMs demonstrated particularly favorable overall survival, probably due to early detection. Older age, Caucasian race, and smoking history were significant risk factors for developing SPM. Analysis of cytogenetics and MSK Impact sequencing analysis remains in progress and will be reported.