KEY POINTS
- This retrospective single-institution study from UT Southwestern Medical Center evaluated 163 lesions in 46 patients with indolent primary cutaneous B-cell lymphoma treated with focal low-dose radiotherapy between 2016 and 2025. The cohort included 84 primary cutaneous follicle center lymphoma lesions (51.5%) and 79 primary cutaneous marginal zone lymphoma lesions (48.5%). Median follow-up was 15 months (IQR 8–27). The primary endpoint was freedom from treatment failure (FFTF), defined as the absence of subsequent radiotherapy or medical retreatment directed at the same lesion.
- Most lesions received the ultra-low-dose 4-Gy regimen. Of 163 lesions, 137 (84%) received 4 Gy, generally as 2 Gy × 2, whereas 17 received 8 Gy and nine received 12 Gy, using several fractionation schedules. Superficial x-ray treatment, typically 80–100 kVp, was used for 151 lesions (92.6%); ten received electrons and two IMRT. Treatment covered visible or palpable disease with an approximately 1–2-cm margin. Dose selection was nonrandomized and could reflect lesion characteristics, treatment history, and physician preference.
- In the overall cohort, higher doses did not significantly improve freedom from retreatment. One-year FFTF was 77% after 4 Gy (95% CI 70–85%) versus 86% after 8–12 Gy (95% CI 73–100%), with p=0.55. However, among 129 previously unirradiated lesions, one-year FFTF was 75% after 4 Gy (n=116; 95% CI 67–84%) versus 100% after 8–12 Gy (n=13; p=0.044). The absolute difference was 25 percentage points, but the higher-dose subgroup contained only 13 lesions and no retreatment events.
- Adjusted analysis also favored 8–12 Gy in previously unirradiated lesions, although uncertainty remained substantial. A Firth penalized Cox model adjusting for lesion stage and anatomical site found lower retreatment hazard with the higher-dose approach (HR 0.093, 95% CI 0.00073–0.68; p=0.012). The extremely wide confidence interval reflects the small event count and complete absence of failures in the higher-dose subgroup. The result should therefore be considered an exploratory association rather than a validated comparative treatment effect.
- Initial objective response was high with both doses, while complete response differed in the de novo subgroup. At approximately three months, overall response rates were 91.7% with 4 Gy versus 100% with 8–12 Gy (p=0.37) among evaluable lesions. For previously unirradiated lesions, corresponding objective response rates were 96.1% versus 100% (p=1.0), while complete response rates were 57.8% versus 91.7% (p=0.027). These findings suggest that initial response and durability of lesion control may not be interchangeable endpoints.
- Anatomical site and previous treatment were also associated with the need for retreatment. One-year FFTF was 69% for head and neck lesions compared with 96% for upper-extremity lesions, with significant anatomical variation across sites (p=0.011). In multivariable analysis, head and neck location was associated with higher retreatment hazard than upper-extremity location (HR 8.11, 95% CI 1.66–39.65; p=0.0097). Previously unirradiated lesions also had better FFTF than reirradiated lesions (HR 0.51, 95% CI 0.27–0.97; p=0.042), suggesting that treatment history may influence apparent dose effectiveness.
- Treatment-related toxicity was low, but the study had important statistical limitations. Only three lesions (1.8%) developed grade 1 radiation dermatitis, while 41 lesions (25.1%) had documented hyperpigmentation or localized alopecia; no grade ≥2 radiation toxicity was reported. Nevertheless, 163 lesions originated from only 46 patients, and the primary statistical analyses treated lesions as independent observations without accounting for within-patient clustering. Retreatment decisions were also clinician-dependent rather than based on centrally reviewed progression criteria, and the median follow-up was too short for firm conclusions about long-term durability.
CLINICAL TAKEAWAY
Ultra-low-dose 4-Gy radiotherapy achieved high initial response rates in indolent primary cutaneous B-cell lymphoma, but approximately one-quarter of previously unirradiated lesions required retreatment within one year. Higher doses of 8–12 Gy were associated with more durable lesion control and similarly limited observed toxicity, although the small and nonrandomized comparison prevents establishing an optimal palliative dose. Prospective studies accounting for patient-level clustering and treatment history are needed before routinely replacing 4 Gy with higher prescriptions.