EOMES+CD45RO+ effector memory T-cells determine response to combined anti-PD-1/anti-CTLA-4 immunotherapy
Tuba Gide (Melanoma Institute Australia, Sydney, New South Wales, Australia) presented a comprehensive analysis of a total of 158 baseline and early during treatment biopsies from 120 patients with advanced melanoma to evaluate potential predictive biomarkers. Three major analyses were performed: RNA sequencing, multiplex immunofluorescence, and mass cytometry (CyTOF). A number of findings were reported which validated the consensus that patients with inflamed tumors are more likely to benefit from anti-PD1 therapy as a single-agent or in combination with an anti-CTLA4 inhibitor. For instance, the presence of a specific subset of T cells, identified as EOMES+CD69+CD45RO+, was found to be associated with improved outcomes in response to combined therapy (CD4+ P=0.040; CD8+ P=0.038). Interestingly, the potential predictive value of cell detection, as measured by receiver operating curve generation, exceeded that of the presence of PDL1 positivity and CD8+ T cells. The application of these findings will require prospective validation prior to adoption of these tests in clinical practice.
Prognostic and predictive value of AJCC-8 staging in the phase 3 EORTC 1325/KEYNOTE-054 trial of pembrolizumab vs placebo in resected high-risk stage III melanoma
Caroline Robert (Gustave Roussy Cancer Centre, Villejuif, France) discussed a retrospective analysis of the KEYNOTE-054 trial, which randomized patients with Stage III melanoma to either 1 year of pembrolizumab or placebo. Here, data were presented based on a recategorization of patients from the AJCC 7th edition, which was in use at the time of trial accrual, to the AJCC 8th edition, which went into routine practice beginning January 2018. Overall, no significant changes occurred between updates; e.g. pembrolizumab is associated with improved outcomes compared with placebo regardless of the AJCC version used to stage patients. However, there were demonstrated changes to the subgroup analysis datasets. KEYNOTE-054 was designed to exclude Stage III patients with the lowest risk of recurrence, namely those with microscopic involvement of one sentinel lymph node with the greatest deposit less than 1 mm. In the reanalysis, the patients with the lowest risk strata (AJCC 8th version Stage IIIA) had an excellent prognosis whether treated with pembrolizumab or placebo. For this reason, this report suggests that Stage IIIA patients as assessed by the AJCC 8th version may not need adjuvant anti-PD1 therapy.
5-year outcomes in patients with ipilimumab-refractory melanoma treated with pembrolizumab in KEYNOTE-002
Antoni Ribas (Jonsson Comprehensive Cancer Center, University of California, Los Angeles, CA, USA) discussed long-term follow up results of KEYNOTE-002 which led to the original FDA-approval of pembrolizumab for patients with advanced melanoma previously treated with ipilimumab. This study randomized patients to one of two doses of pembrolizumab (2 mg/kg and 10 mg/kg every three weeks) or to investigator’s choice chemotherapy in a 1:1:1 fashion. Significantly, 5-year outcomes revealed that both the overall response rate and complete response rate of patients treated with pembrolizumab (data from patients treated at both doses were pooled) improved (27.4 up from 22.7% and 8.3 up from 2.8%, respectively). Additionally, the majority of patients with either a complete response or partial response have remained in response with nearly five years of median follow up. Furthermore, the three and four year progression free and overall survival (PFS, OS) were as follows: PFS 36 months: 16.9%, 48 months: 15%; OS 36 months: 30%, 48 months: 27.2%. In summary, this trial, which accrued Stage IV melanoma patients treated with single-agent ipilimumab prior to anti-PD1 therapy, provides further evidence from a well annotated dataset suggestive of the long-term durable benefit of pembrolizumab, even in patients with a poor prognosis.
Long-term follow-up of standard-dose pembrolizumab plus reduced-dose ipilimumab in 153 patients with advanced melanoma: KEYNOTE-029 1B
Georgina Long (Melanoma Institute Australia, Sydney, New South Wales, Australia) detailed long-term follow up data from 153 patients enrolled in the 1B cohort of the KEYNOTE-029 study, which tested the combination of pembrolizumab and low-dose ipilimumab in patients with advanced melanoma. In this update, which includes two more years of data, initial findings suggested that treatment with standard-dose pembrolizumab plus low-dose ipilimumab is well tolerated and effective. Specifically, originally reported response rate, complete response rate, and grade 3 or higher toxicity rates improved from 61%, 15%, and 45% respectively, to 62%, 27%, and 47%. Importantly, two and three year duration of response (DOR), progression free survival (PFS) and overall survival (OS) reports also reported positive outcomes (DOR: 2 yr 87%, 3yr 84%; PFS: 2yr 61%, 3yr 59%; OS: 2 yr 79%, 3yr 73%). While such outcomes demonstrate high levels of therapeutic efficacy compared to other dual anti-PD1/anti-CTLA-4 therapies, these findings are still preliminary with respect to adoption in clinical practice. In particular, the patient population was enriched for “better” prognosis patients (e.g. only 25% had an elevated LDH) and a randomized trial should be reported before this data is widely used. Nonetheless, this approach appears feasible and associated with promising efficacy.
IFNγ-Activated dermal lymphatic vessels inhibit cytotoxic CD8+ T cells in melanoma
Amanda Lund (Oregon Health and Science University, Portland, OR, USA) presented a new concept that is based on the following hypothesis: lymphatic vessel changes in the course of cancer progression block the immune response by physically impairing T-cell infiltration. Lund’s group demonstrated, in preclinical syngeneic mouse melanoma models (B16, YUMM1.7, YUMMER1.7) that non-hematopoietic, non-tumor PD-L1 expression, primarily by lymphatic and blood endothelial cells, limits CD8+ T cell accumulation in melanoma. Interestingly, IFNγ produced by tissue infiltrating, antigen-specific CD8+ T cells was determined sufficient to activate lymphatic vessel PD-L1 expression and essentially lead to a negative feedback loop that prevented further T-cell infiltration; disruption of IFNγ-dependent signaling through loss of IFNγR in the lymphatic system boosted T cell accumulation in melanoma and lead to CD8+ T cell-mediated tumor control and improved overall survival. The importance of this finding is the identification of the lymphatic vasculature as a dynamic factor in the immune microenvironment. This system may prove to be an important target to enhance the effectiveness of anti-cancer immunotherapy.
Fecal microbiota transplantation in metastatic melanoma patients who failed immunotherapy - Preliminary results from a microbiome-based phase I clinical trial
Erez Baruch (Ella Lemelbaum Institute for Immuno-Oncology, Sheba Medical Center, Tel-HaShomer, Israel and Department of Clinical Immunology and Microbiology, Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel) reported on a clinical trial (NCT03353402) which used fecal microbiota transplantation (FMT) to enhance clinical response to anti-PD-1 therapy in melanoma patients who have previously failed all standard lines of treatment. This study is the first of many launched to present preliminary data following a number of reports demonstrating that the gut microbiota may influence responsiveness to anti-PD1 therapy. In this trial, FMT donors were metastatic melanoma patients who previously had a durable complete response to anti-PD-1 therapy. FMT was delivered through two routes: colonoscopy and stool capsules. Collected pre and post-FMT, samples included stools for 16S rRNA gene sequence and metagenomic analyses, serum for cytokines measurements, peripheral blood mononuclear cells for RNA sequencing as well as gut and tumor biopsies for histological analyses and RNA sequencing. Data from the first five patients were presented, four of whom had primary resistance to anti-PD1 therapy. In these patients, no safety issues were identified and there was a clear change in the microbiota of the patients enrolled, and in one, the recipient (enrolled patient) had stool containing an extended spectrum beta-lactamase (ESBL) producing species that was cleared when the donor (ESBL negative) stool engrafted. The efficacy outcomes of the first three patients were available – 2 had primary progression and 1 had improvement of disease-related symptoms and mild reduction of tumors. Preliminary pharmacodynamic/correlative studies were minimal to date, although there was some suggestion of immune responsiveness. In summary, FMT in the setting of heavily treated metastatic melanoma patients demonstrated an acceptable safety profile. Further studies combining FMT with anti-PD-1 therapy will shed light on prospective clinical response.
A role for NK cells in clinical response of stage IV melanoma patients treated with anti-PD-1
Hansol Lee (Melanoma Institute Australia, North Sydney, New South Wales, Australia) discussed her group’s analysis of tumor samples from melanoma patients treated with anti-PD1 therapy with a focus on the relationship between natural killer (NK) cells and clinical response. In all, 45 metastatic melanoma patients treated with the checkpoint inhibitor were categorized by responders (complete response (CR)/partial response (PR)/stable disease (SD)>6 months) and non-responders (SD <6 months/progressive disease (PD)). Three major analyses were performed on tissue samples prior to treatment and on a subset of tumor samples from early during treatment: whole transcriptome sequencing, multiplex immunofluorescent staining and spatial distribution analysis. Overall, there were 9 NK cell-specific genes found to be upregulated and a significantly higher density of intra- and peri-tumoral CD16+ and granzyme B+ NK cells in responders versus non-responders. Additionally, NK cells were observed closer in proximity to tumor cells in responders vs non-responders, and PD-1 blockade enhanced NK cell cytotoxicity, suggesting that NK cells may play a key role in mediating response to anti-PD-1 therapy.