Background: Advanced squamous cell anal carcinoma (advSCAC) is a rare disease with poor prognosis. No standard therapies beyond first line are currently available, yet a promising activity was documented for the anti-EGFR cetuximab (CET) and for anti-PD-1 agents in previous retrospective case series and phase I-II studies, respectively. In experimental models combination of EGFR and PD-L1 blockade was synergistic as PD-L1 blockade led to NK cells activation enhancing cetuximab ADCC. In this trial we aimed to evaluate safety and activity of the anti-PD-L1 avelumab (AVE) alone or in combination with CET in pretreated advSCAC. Methods: This was an open-label, prospective, multicenter randomized phase 2 trial (NCT03944252). Patients (pts) with advSCAC progressed after at least 1 line of treatment were randomized 1:1 to receive either AVE 10 mg/kg (arm A) or AVE + CET 500 mg/sqm (arm B) as bi-weekly regimens. A Simon’s two-stage Mini-Max design was used. The null hypothesis of a true response rate 5% was tested against the one-sided alternative of a true response rate 20% in each arm. Setting type I error at 0.05 and power at 80%, 30 pts per arm had to be randomized. No formal comparison between the two arms was planned. Primary endpoint was overall response rate (ORR); secondary endpoints were Progression-Free Survival (PFS), Overall Survival (OS) and safety. Results: Sixty pts were enrolled, 30 in each arm. All baseline characteristics were well balanced between the two arms. Median age was 63 years; M/F was 19/41; 12 out of 30 pts in each arm had distant metastases; 7 in arm A and 10 in arm B received > 1 previous lines of treatment. At a median follow up of 8.7 months, 3 out of 30 pts in each arm obtained PR (ORR 10%); SD was observed in 12 pts in arm A (40%) and 14 in arm B (47%). Disease control rate was thus 50% in arm A and 57% in arm B. Duration of disease control was 6.1 (95%CI 3.7–11.0) and 6.1 (95%CI 4.1–9.6) months in arm A and B, respectively. Median PFS was 2.1 (95%CI 1.8–4.0) in arm A and 3.9 months (95%CI 2.1–5.6) in arm B. Grade 3-4 adverse events were 13.3% in arm A and 33.3% in arm B: anemia 10% vs 13.3%, fatigue 0 vs 6.7%, skin toxicity 0 vs 6.7%. Treatment interruption due to AE occurred in 3 pts, 1 in arm A and 2 in arm B. Translational analyses will be performed on tissue and blood samples for exploratory purpose. Conclusions: The CARACAS trial was the first clinical study to test dual EGFR and PD-L1 blockade in advSCAC. Both AVE monotherapy and AVE-CET showed promising activity with manageable safety profile. Clinical trial information: NCT03944252.