Predictors of Local Control with Palliative Radiotherapy for Multiple Myeloma

Abstract

Palliative radiotherapy (RT) is employed for patients with multiple myeloma to improve or prevent symptoms. However, the optimal dose fractionation is not well defined. The role of cytogenetics in informing RT warrants further study. We performed an institutional analysis of patients with multiple myeloma receiving palliative RT and assessed factors associated with local progression, with a focus on dose fractionation and cytogenetic abnormalities. We queried a prospectively maintained, departmental database for consecutive patients who received palliative RT for multiple myeloma at our institution from 2015 to 2020. Double- and triple-hit were defined as the presence of two and three high-risk cytogenetic abnormalities. RT dose fractionation data were extracted from the database. Follow-up imaging was used to evaluate for progression. A total of 239 patients with 362 treated lesions were included. Twenty-five patients (10.4%) with 39 lesions had double-hit cytogenetics, and 4 patients (1.7%) with 7 lesions were triple-hit. Patients had the following number of lesions treated with RT: 1 (156, 65.3%), 2 (53, 22.2%), 3 (17, 7.1%), or >3 (13, 5.4%). The most commonly targeted sites were spine (125, 34.5%), abdomen/pelvis (67, 18.5%), and lower extremity (53, 14.6%). Most lesions received doses of 20 Gy/5 fx (132, 36.5%), 8 Gy/1 fx (93, 25.7%), or 30 Gy/10 fx (48, 13.3%). RT equivalent dose in 2 Gray fractions (EQD2) was <2000 cGy for 126 lesions (34.8%) and ≥2000 cGy for 236 lesions (65.2%). At a median follow-up of 4.3 years, the risk of local progression on a per lesion basis at 1 and 4 years was 7.8% (95% CI: 5.5-11.1) and 13.4% (10.3-17.5), respectively. No cytogenetic abnormalities were correlated with local progression. Factors significant on univariate analysis included female sex [hazard ratio (HR): 1.94 (1.02-3.71), p = .045], LDH at diagnosis [HR per 10 units/liter: 1.04 (1.09-1.08), p = .016], and number of treated lesions [HR per lesion: 1.38 (1.02-1.89), p = .039]. These three covariates were included on multivariable analysis, and the only covariate to approach significance was number of treated lesions [HR for >3 versus 1: 2.43 (0.88-6.74), p = .059]. In the overall cohort, EQD2 did not impact risk of progression. Among those with >3 treated lesions, EQD2 ≥2000 cGy was associated with a significantly lower risk of progression [HR: 0.05 (0.01-0.23), p<.001]. Double- and triple-hit status were not correlated with progression. Median overall survival in all patients was 4.1 years versus 1.5 and 0.6 years in those with double- and triple-hit disease, respectively. In this large, institutional study of patients with multiple myeloma, palliative RT achieves durable long-term local control. Patients with high disease burden may be at increased risk of progression at treated sites. This group may benefit from an EQD ≥2000 cGy. Cytogenetics, including double- and triple-hit status, do not appear to influence RT response.