Risk for Second Primary Hematologic Malignancies by Radiotherapy Technique in Prostate Cancer Survivors

Abstract

<h3>Purpose/Objective(s)</h3> Among the most serious adverse effects after radiotherapy for prostate cancer is the development of second malignant neoplasms. Recent advances in radiotherapy techniques, such as the introduction of intensity-modulated radiation therapy (IMRT) and proton beam radiotherapy (PBRT), have aimed to reduce exposure to adjacent healthy tissues but may lead to greater bone marrow exposure. Risks for developing myelodysplastic syndrome/acute myeloid leukemia (MDS/AML) but not lymphoid malignancies after radiotherapy have been demonstrated. We quantified risks of hematologic malignancies following IMRT and PBRT compared to 3D conformal radiotherapy (CRT) in a large cohort of prostate cancer patients. <h3>Materials/Methods</h3> This retrospective cohort study used the linkage of Surveillance, Epidemiology, and End Results (SEER) cancer registry data with Medicare claims, including men who were diagnosed with first primary non-metastatic prostate cancer at ages 66-84 years from 2002-2014, received external beam radiotherapy (IMRT, PBRT, or CRT) but no chemotherapy in the first year after diagnosis, and survived ≥2 years after diagnosis without a second primary cancer. Using Cox regression, hazard ratios (HR) and corresponding 95% confidence intervals (CI) were computed to estimate risk of second primary hematologic malignancies, classified into subtypes according to the World Health Organization, after IMRT and PBRT vs CRT. Models were adjusted for age at prostate cancer diagnosis, tumor grade, race, Charlson comorbidity score, and receipt of initial prostate cancer therapy. <h3>Results</h3> The cohort (median follow-up=6.8 years) included 69,966 patients, of whom 23,806 received CRT alone, 43,901 received IMRT without PBRT, and 2,259 received PBRT. A total of 387 patients developed a myeloid malignancy, most commonly MDS/AML, N=291), and 739 developed a lymphoid malignancy, most commonly plasma cell neoplasms (N=178) and diffuse large B-cell lymphoma (N=159). Compared to patients who received CRT, neither IMRT nor PBRT was associated with risk of developing myeloid malignancies overall (IMRT: N=211, HR=0.89, 95%CI 0.70-1.13; PBRT: N=13, HR=0.97, 95%CI 0.55-1.74), with similar results for MDS/AML specifically. Similarly, neither IMRT nor PBRT was associated with risk of developing lymphoid malignancies overall (IMRT: N=390, HR=0.93, 95%CI 0.78-1.11; PBRT: N=24, HR=0.94, 95%CI 0.61-1.43) compared to CRT, with similar results for specific lymphoid malignancy subtypes. <h3>Conclusion</h3> This is among the first cohort studies analyzing risk of hematologic malignancies by subtype after prostate cancer radiotherapy. With longer follow-up and larger sample size compared to previous reports, the hypothesized change in risk for MDS/AML with IMRT/PBRT compared to CRT was not observed. As more data accumulates from patients receiving PBRT, we can gain greater insight into the technique's long-term effects.