Randomized phase II trial of pembrolizumab/lenvatinib (P+L) +/- responder fecal microbiota transplant (R-FMT) in PD-1 relapsed/refractory (R/R) cutaneous melanoma (MEL).

TPS9607 Background: A significant portion of patients (pts) with advanced MEL develop resistance to immune checkpoint inhibitors (ICI). The gut microbiome is a major tumor-extrinsic regulator of ICI response. In mice, gut microbiota modulate therapeutic activity of ICI, and administration of certain gut commensals or responder-derived fecal microbiota transplantation (R-FMT) promotes anti-PD-1 efficacy in melanoma-bearing mice. In the LEAP-004 trial, P+L produced 21% ORR in PD-1 R/R MEL with Gr3+ TRAE 45.6%. Longitudinal analyses of gut microbiome samples from ICI-treated cancer pts suggests key species including Actinobacteria and Firmicutes are associated with favorable response. We and others have previously demonstrated that microbiome modulation using R-FMT reversed PD-1 non-response in PD-1 R/R MEL. In PD-1 naïve MEL, healthy donor FMT (hdFMT) augments benefit of PD-1 monotherapy. We hypothesized that intestinal dysbiosis mediates PD-1 resistance in PD-1 R/R MEL, microbiome modulation may reverse TKI toxicity, and that R-FMT may resensitize MEL to PD-1 blockade. Methods: This is an investigator-initiated, randomized, phase II trial of P+L vs. P+L+R-FMT in PD-1 R/R MEL. Pts must have measurable disease per RECIST v1.1, no active CNS metastases (or if present, previously treated and stable on repeat imaging), and no contraindications to FMT administration. R-FMT is derived from non-obese, advanced MEL with durable response (mPFS ≥24 months) following prior PD-1 ICI, with no recent (≤1 month) antibiotic exposure. Donors are screened broadly for infectious agents including SARS-CoV-2 using bookend screening. R-FMT is processed to make fecal infusates and orally bioavailable capsules. Eligible pts are randomized 1:1 to either P+L P 200mg every 3 weeks (Q3W) along with L 20mg orally once daily or P+L+R-FMT. R-FMT is administered endoscopically on D1, and D42, and thereafter via capsules until progression or unacceptable toxicity. Response is assessed at D73, and thereafter Q12W. Pts undergo biopsies at baseline and W8, with optional biopsy at progression. Primary endpoint: ORR assessed using RECIST v1.1 by Blinded Independent Central Review (BICR). Secondary endpoints include: safety, progression-free survival (PFS), overall survival (OS), and landmark PFS/OS. Key translational endpoints include immune activation (tumor tissue, blood), metagenomic engraftment, and transcriptomic analyses of intestinal luminal myeloid cells and exfoliome. The null hypothesis that the true ORR is 20% will be tested versus a one-sided alternative hypothesis of ≥50%. This design has a type I error rate 0.10 and power 0.8 when the true response rate is 0.25. Enrollment has commenced. Clinical trial information: NCT06030037 .