New fecal microbiota for cancer patients: The composition of the gut microbiome influences the response of cancer patients to immunotherapies. Baruch et al. and Davar et al. report first-in-human clinical trials to test whether fecal microbiota transplantation (FMT) can affect how metastatic melanoma patients respond to anti–PD-1 immunotherapy (see the Perspective by Woelk and Snyder). Both studies observed evidence of clinical benefit in a subset of treated patients. This included increased abundance of taxa previously shown to be associated with response to anti–PD-1, increased CD8 ⁺ T cell activation, and decreased frequency of interleukin-8–expressing myeloid cells, which are involved in immunosuppression. These studies provide proof-of-concept evidence for the ability of FMT to affect immunotherapy response in cancer patients. Science , this issue p. 602 , p. 595 ; see also p. 573

Background: Immunotherapy has drastically changed the outlook for melanoma patients over the past decade. Specifically, the dual blockade of immune checkpoints using ipilimumab and nivolumab has shown unprecedented response rates and survival outcomes. This immense achievement, though, is at the cost of toxicity, with 60% of the patients experiencing high-grade adverse events (AEs). Our study aims to report the efficacy and toxicity outcomes of an out-of-trial, real-life population. Methods: Data on metastatic melanoma patients treated with ipilimumab and nivolumab were retrieved from our melanoma database—a single-center prospectively updated, medical-records based oncologic registry. Data included demographics, clinical and pathological information, as well as tumor responses and survival. Associations between patient or treatment characteristics and outcomes were also evaluated. Results: We identified 172 metastatic melanoma patients, of whom 64% were treatment-naïve. The median follow-up was 12 months. The response rates for treatment-naïve and previously-treated patients were 61% and 25%, respectively; median progression-free survival (PFS) were 12.2 and 2.6 months, and median overall survival (OS) were not-reached (NR) and 6.1 months, respectively. The estimated three-year OS for treatment-naïve patients was 58% (95% CI 42–65). At data cutoff, 22% were still on-treatment. Grade 3–4 adverse events (AEs) were reported in 60% of the patients, almost all of whom were exposed to steroid treatments (59%); AEs were fatal in 4 patients, and led to permanent treatment discontinuation in 31%. Factors significantly associated with outcome were cutaneous histology, low lactate dehydrogenase (LDH), low number of metastatic sites, performance status, first line of treatment and number of combinations administered during the induction phase. Conclusions: Despite the profoundly different baseline patient characteristics, the combination of ipilimumab and nivolumab is as effective in the real-world population as it was in clinical trials, including long-term outcomes. In addition to confirming the significance of baseline prognostic factors, our study reveals that the number of combinations effectively administered may also be correlated with good outcome.

Background: Adoptive cell therapy with T cells genetically engineered to express a chimeric antigen receptor (CAR-T) or tumor-infiltrating T lymphocytes (TIL) demonstrates impressive clinical results in patients with cancer. Lymphodepleting preconditioning prior to cell infusion is an integral part of all adoptive T cell therapies. However, to date, there is no standardization and no data comparing different non-myeloablative (NMA) regimens. Methods: In this study, we compared NMA therapies with different doses of cyclophosphamide or total body irradiation (TBI) in combination with fludarabine and evaluated bone marrow suppression and recovery, cytokine serum levels, clinical response and adverse events. Results: We demonstrate that a cumulative dose of 120 mg/kg cyclophosphamide and 125 mg/m² fludarabine (120Cy/125Flu) and 60Cy/125Flu preconditioning were equally efficient in achieving deep lymphopenia and neutropenia in patients with metastatic melanoma, whereas absolute lymphocyte counts (ALCs) and absolute neutrophil counts were significantly higher following 200 cGyTBI/75Flu-induced NMA. Thrombocytopenia was most profound in 120Cy/125Flu patients. 30Cy/75Flu-induced preconditioning in patients with acute lymphoblastic leukemia resulted in a minor ALC decrease, had no impact on platelet counts and did not yield deep neutropenia. Following cell infusion, 120Cy/125Flu patients with objective tumor response had significantly higher ALC and significant lower inflammatory indexes, such as neutrophil-to-lymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR). Receiver-operating characteristics curve analysis 7 days after cell infusion was performed to determine the cut-offs, which distinguish between responding and non-responding patients in the 120Cy/125Flu cohort. NLR≤1.79 and PLR≤32.7 were associated with clinical response and overall survival. Cytokine serum levels did not associate with clinical response in patients with TIL. Patients in the 120Cy/125Flu cohort developed significantly more acute NMA-related adverse events, including thrombocytopenia, febrile neutropenia and cardiotoxicity, and stayed significantly longer in hospital compared with the 60Cy/125Flu and TBI/75Flu cohorts. Conclusions: Bone marrow depletion and recovery were equally affected by 120Cy/125Flu and 60Cy/125Flu preconditioning; however, toxicity and consequently duration of hospitalization were significantly lower in the 60Cy/125Flu cohort. Patients in the 30Cy/75Flu and TBI/75Flu groups rarely developed NMA-induced adverse events; however, both regimens were not efficient in achieving deep bone marrow suppression. Among the regimens, 60Cy/125Flu preconditioning seems to achieve maximum effect with minimum toxicity.

Immune-checkpoint inhibitor (ICI)–related diarrhea is attributed to inflammatory colitis, with no other drug-related differential diagnosis. Here, we investigated the occurrence of pancreatic atrophy (PA) in ICI-treated cancer patients and its correlation to exocrine pancreatic insufficiency (EPI). Metastatic melanoma, non–small cell lung carcinoma, and head and neck squamous cell carcinoma patients (n = 403) treated with anti–PD-1 (n = 356) or anti–CTLA-4 (n = 47) were divided into a case group (radiologic evidence of PA); control group matched by age, gender, and previous lines of treatment; and colitis group (ICI-induced colitis). Quantitative pancreatic volumetry was used for calculation of the decrease in pancreatic volume over time (atrophy rate). Thirty-one patients (7.7%) developed PA compared with 41 matched controls (P = 0.006). Four patients developed EPI, all from the anti–PD-1–treated group, which resolved with oral enzyme supplementation. The atrophy rate did not correlate with EPI (P = 0.87). EPI-related diarrhea presented at a median of 9 months, whereas the diarrhea of anti–PD-1–induced colitis patients (n = 22) was presented at a median of 2 months (P = 0.029). ICI-induced PA is irreversible and can result in EPI. EPI should be suspected in cases of late-onset steroid-resistant diarrhea with features of steatorrhea and treated with oral enzyme supplements.