To characterize local relapse after surgical fixation and postoperative radiotherapy (RT) for multiple myeloma (MM) with cortical involvement of long bones.

We retrospectively identified patients with MM involving cortical long bones treated with surgical fixation followed by postoperative RT at our institution. Local failures, defined as radiographic recurrence along the surgical hardware, were documented, and potential associations of independent variables (RT dose, fractionation, and extent of hardware coverage) with local failure were assessed by univariate Cox regression.

We identified 33 patients with 40 treated sites with a median follow-up of 25.7 months; 68% of treatments were for pathologic fracture, and 32% were for impending fracture. The most common dose and fractionation were 20 to 25 Gy in 8 to 12 fractions. On average, 76% of the surgical hardware was covered by the postoperative RT field (median, 80%; range, 28%-100%). Local failure was observed in 5 cases (12.5%), 2 within the RT field and 3 out of field. None of the relapses resulted in hardware failure, and 2 were retreated with RT. The extent of hardware coverage predicted disease relapse along the hardware (hazard ratio = 6.44; 95% confidence interval, 1.09-37.97; P = .04); however, total RT dose, biologically effective dose, and number of fractions did not.

After internal fixation of long bones with MM, full hardware coverage with the RT field could reduce the risk, though small, of disease developing in the future in the proximate hardware. Postoperative RT doses of 20 to 25 Gy in 8 to 10 fractions can achieve excellent local control.