Abscopal effects in patients with malignant melanoma treated with radiotherapy and immune checkpoint inhibition: analysis of a large observational multicenter study.

Background: Abscopal effect (AbE), the regression of non-irradiated metastatic lesions (NILs) following radiotherapy (RT), is relevant in patients with malignant melanoma (MM) with progressive disease (PD) under immune checkpoint inhibition (ICI) as resistance to immunotherapy. In the "ARTIC" trial, we assessed the incidence of AbE in patients with progressive MM by evaluating the effect of RT on NILs.

Methods: ARTIC (Abscopal effects in metastasized cancer patients treated with RadioTherapy and Immune Checkpoint inhibition) (ARO (Arbeitsgemeinschaft Radiologische Onkologie) 2022-10, DRKS00032390) retrospectively screened clinical records of patients with stage IV MM with PD under ICI. Patients received RT for metastases and had ≥1 NIL outside the RT field (=control lesion). NILs were evaluated according to iRECIST (immune Response Evaluation Criteria in Solid Tumors): abscopal response (AR): size reduction ≥30%, abscopal progression (AP): size increase ≥20%, abscopal control (AC): all others. Patients with AR and/or AC were categorized as abscopal benefit (AB), patients with AP and/or mixed response=no AB. RT details and factors influencing AR were analyzed.

Results: After screening clinical records of 3773 patients with stage IV tumor from 12 oncological centers in Germany, we identified 47 patients with MM with 115 NILs. RT targeted metastases in brain (38.3%) and lung (19.1%), primarily using stereotactic RT (29.8%). The mean time interval between the end of ICI and RT was 3.53±5.67 months. AR was achieved in 19.1% of patients and 29.1% of lesions. Compared with stereotactic RT, normofractionated or other (non-stereotactic) RT regimens significantly reduced the probability of AB (OR=0.092, p=0.04, 95% CI: (0.007 to 0.758)). Longer ICI-to-RT intervals were associated with reduced mortality risk (HR=0.703, p=0.007, 95% CI: (0.544 to 0.908)). Patients with AB had a longer median overall (17 vs 9 months) and a longer median progression-free survival (4 vs 2 months).

Conclusions: RT can induce AR in patients with MM with PD under ICI, particularly with hypofractionated regimens and long ICI-to-RT intervals. Our findings can serve as a reference for designing prospective trials.