Second primary malignancies with lenalidomide therapy for newly diagnosed myeloma: A meta-analysis of individual patient data

Background: Lenalidomide has been linked to second primary malignancies in myeloma. We aimed to pool and analyse available data to compare the incidence of second primary malignancies in patients with and without lenalidomide exposure. Methods: We identified relevant studies through a search of PubMed and abstracts from the American Society of Clinical Oncology, American Society of Hematology, and the International Myeloma Workshop. Randomised, controlled, phase 3 trials that recruited patients with newly diagnosed multiple myeloma between Jan 1, 2000, and Dec 15, 2012, and in which at least one group received lenalidomide were eligible for inclusion. We obtained individual patient data (age, sex, date of diagnosis, allocated treatment and received treatment, duration of treatment and cause of discontinuation, maintenance treatment, date of first relapse, date of second primary malignancy diagnosis, type of second primary malignancy, date of death or last contact, and cause of death) by direct collaboration with the principal investigators of eligible trials. Primary outcomes of interest were cumulative incidence of all second primary malignancies, solid second primary malignancies, and haematological second primary malignancies, and were analysed by a one-step meta-analysis. Findings: We found nine eligible trials, of which seven had available data for 3254 patients. 3218 of these patients received treatment (2620 had received lenalidomide and 598 had not), and were included in our analyses. Cumulative incidences of all second primary malignancies at 5 years were 6·9% (95% CI 5·3-8·5) in patients who received lenalidomide and 4·8% (2·0-7·6) in those who did not (hazard ratio [HR] 1·55 [95% CI 1·03-2·34]; p=0·037). Cumulative 5-year incidences of solid second primary malignancies were 3·8% (95% CI 2·7-4·9) in patients who received lenalidomide and 3·4% (1·6-5·2) in those that did not (HR 1·1 [95% CI 0·62-2·00]; p=0·72), and of haematological second primary malignancies were 3·1% (95% CI 1·9-4·3) and 1·4% (0·0-3·6), respectively (HR 3·8 [95% CI 1·15-12·62]; p=0·029). Exposure to lenalidomide plus oral melphalan significantly increased haematological second primary malignancy risk versus melphalan alone (HR 4·86 [95% CI 2·79-8·46]; p<0·0001). Exposure to lenalidomide plus cyclophosphamide (HR 1·26 [95% CI 0·30-5·38]; p=0·75) or lenalidomide plus dexamethasone (HR 0·86 [95% CI 0·33-2·24]; p=0·76) did not increase haematological second primary malignancy risk versus melphalan alone. Interpretation: Patients with newly diagnosed myeloma who received lenalidomide had an increased risk of developing haematological second primary malignancies, driven mainly by treatment strategies that included a combination of lenalidomide and oral melphalan. These results suggest that alternatives, such as cyclophosphamide or alkylating-free combinations, should be considered instead of oral melphalan in combination with lenalidomide for myeloma. Funding: Celgene Corporation.