DECEMBER 2020
The following articles have been recommended for further reading in the field of cancer immunotherapy by JITC's Clinical Trials Monitor Section Editor, Leisha A. Emens, MD, PhD.
“Neoadjuvant PD-L1 plus CTLA-4 blockade in patients with cisplatin-ineligible operable high-risk urothelial carcinoma” by Jianjun Gao et al
Cisplatin-based neoadjuvant chemotherapy provides long-term survival benefits in patients with muscle-invasive urothelial carcinoma (MIUC), but as many as 50% of patients with MIUC cannot receive cisplatin-containing chemotherapy. Between April 2017 and December 2018, Jianjun Gao and colleagues enrolled 28 patients with high-risk, cisplatin-ineligible, localized MIUC on the first pilot trial of combination neoadjuvant anti-PD-L1 plus anti-CTLA-4 (NCT02812420). Each patient had baseline transurethral resection of bladder tumor followed by combination treatment with durvalumab (1,500 mg/kg) and tremelimumab (75 mg/kg) every 4 weeks for two cycles before cystectomy. The primary endpoint was safety and the study did not exceed its safety or futility rules. The majority of patients experienced an immune-related adverse event (irAE) of any grade, with the most common being grade 1 or 2 rash and asymptomatic increase in amylase. Grade 3-4 irAEs occurred in 6 of 28 patients, consisting of asymptomatic laboratory abnormalities, hepatitis, and colitis. The pathological complete response (pCR) rate was 37.5% and downstaging to pT1 or less was observed in 58% of patients who completed surgery (n = 24). In the in the 12 patients with large T3/T4 tumors, the pCR rate was 42% and the downstaging rate was 75%. Response rates did not correlate with severity of irAEs, PD-L1 expression on tumor or immune cells, tumor mutation burden (TMB), predicted neoantigens, defects in DNA-damage response genes correlated, nor mutations in KRAS, PIK3CA, PBRM1, EGFR, NRAS, APC2, or FGFR. Response to treatment did correlate with a higher frequency of tertiary lymphoid structures (TLSs) in pre-treatment tumor tissues, and significantly higher densities of B cells, CD4+ T cells and CD8+ T cells in pre-immunotherapy samples. In 16 available pre-treatment samples with TLS, significantly higher expression of POU2AF1 was seen in responder tumors (n = 9) as compared to nonresponders. The data show a tolerable safety profile and encouraging efficacy with two cycles of neoadjuvant combination immunotherapy for cisplatin ineligible patients with MIUC—further trials are needed to determine optimal duration of treatment.
“Overcoming primary and acquired resistance to anti-PD-L1 therapy by induction and activation of tumor-residing cDC1s” by Takaaki Oba et al
Poor intratumoral T cell infiltration is a major mechanism associated with primary and acquired resistance to PD-(L)1 blockade, limiting efficacy of immunotherapy for a substantial proportion of patients. Takaaki Oba and colleagues show that in situ modulation of the dendritic cell compartment in the tumor microenvironment can overcome resistance to PD-(L)1 checkpoint inhibition and demonstrate a pivotal role for conventional type 1 dendritic cells (cDC1s) in shaping the tumor-infiltrating T cell repertoire. In three different murine models of poorly-T cell-inflamed cancers (B16 melanoma as well as AT-3 and 4T-1 triple negative mammary cancers) a combinatorial cDC1-targeted in situ immunomodulation regimen consisting of Flt3L administration, radiotherapy and dual TLR3/CD40 stimulation led to tumor regression and prolonged survival. The triple-combination sensitized tumors to anti-PD-(L)1 therapy. Regression of distal tumors was observed with the regimen, and delivery of all three components to the primary tumor in situ was required for maximal efficacy. In situ immunomodulation resulted in the generation of intratumoral CD44+ CD62L− effector memory T cell subsets and upregulation of PD-1 expression, most prominently on Tet+ CD8+ T cells. The therapeutic effects required myeloid-derived Batf3-dependent cells. T cell trafficking from secondary lymphoid organs was also necessary for antitumor efficacy, along with interferon gamma and IL-12. Single-cell RNA sequencing of myeloid cells revealed a reduced frequency of macrophages in tumors treated with the triple regimen, coinciding with elevated numbers of a cell population with expression profiles consistent with IL-12-producing DCs. In the lymphoid compartment, the regimen led to a massive influx of Slamf6-expressing cells with transcriptomes suggesting a stem-like progenitor-exhausted phenotype. Notably, repeated in situ immunomodulation led to an expansion of novel T cell clones between cycles, leading to durable complete responses and rejection upon rechallenge.
“Mutations in BRCA1 and BRCA2 differentially affect the tumor microenvironment and response to checkpoint blockade immunotherapy” by Robert M. Samstein et al
DNA repair deficiencies may enhance tumor immunogenicity through the generation of neoantigens, but several tumor types with deficiencies in homologous recombination show low responses to checkpoint blockade. By characterizing the immune microenvironment in tumors deficient for BRCA1 or BRCA2, Robert M. Samstein et al filled in some of the gaps for how specific components of homologous recombination repair may shape response to checkpoint blockade. In a pan-cancer cohort of patients profiled with the US FDA-approved MSK-IMPACT targeted capture sequencing assay, somatic or germline truncating mutations in BRCA2 were associated with improved overall survival after checkpoint blockade relative to germline BRCA1 deficiencies. RNA-seq data from the Cancer Genome Atlas (TCGA) breast dataset showed that patients with biallelic inactivating mutations in BRCA1 or BRCA2 formed two distinct clusters, particularly for innate immune cell populations. In both the TCGA and METABRIC datasets, comparison of BRCA1- and BRCA2-deficient tumors to each other and to their nonmutated histological counterparts (to control for known immunological effects of the estrogen receptor), showed that BRCA1-deficient triple-negative tumors were enriched for expression of an immunoregulatory gene set, including several known suppressive and checkpoint proteins such as CD47, CTLA4, CXCL5, ICOS, IDO1, LAG3, PDCD1LG2, SIRPA, TNFRSF9 (4-1BB) and VEGFA. BRCA2-deficient ER+ tumors also had higher numbers of SNVs and indels—indicative of heightened potential immunogenicity—compared to BRCA1-deficient tumors (which tended to have more frequent large rearrangements). Data from mouse models recapitulated several results from human tumors, and also lent further mechanistic insight. 4T-1 tumors (a murine model for triple-negative breast cancer) lacking Brca2 showed delayed growth in response to anti-PD-1 as well as combination anti-PD-1/anti-CTLA-4 therapy, whereas Brca1 null tumors were not responsive to checkpoint blockade—similar patterns were seen in the CT26 colorectal murine model. In serial passage experiments, Brca2-null 4T-1 cells accumulated more SNVs compared to Brca1 knockouts. Comparative RNA-seq analyses revealed enrichment for expression of adaptive and innate immune activation in Brca2-null tumors, with enhanced cytokine signaling, interferon-alpha signaling and natural killer cell-mediated cytotoxicity. Tumor-associated macrophage (TAM) and dendritic cell populations were differentially enriched in Brca1 and Brca2 null tumors, with elevated plasmacytoid dendritic cell populations in Brca2-deficient cells. This is the first study to directly compare the effect of BRCA1 and BRCA2 loss on the tumor microenvironment. Overall, it demonstrates fundamental differences in the impact of BRCA1 and BRCA2 deficiency on the mutational landscape and immune microenvironment of the tumor, and on the tumor response to immune checkpoint blockade.
“Inborn errors of type I IFN immunity in patients with life-threatening COVID-19” by Qian Zhang et al and “Autoantibodies against type I IFNs in patients with life-threatening COVID-19” by Paul Bastard et al
Beyond age, male sex, and the presence of comorbidities, very little is known about why COVID-19 varies so dramatically in clinical course, ranging from completely asymptomatic silent infection to severe and sometimes fatal systemic inflammatory pathology. A pair of papers by Qian Zhang et al and Paul Bastard et al reveal a central role for type I interferon (IFN) signaling in preventing severe disease after SARS-CoV-2 infection. In the first paper, Qian Zhang and colleagues identified rare variants predicted confer loss of function in a set of 13 genes known to control Toll-like receptor 3 (TLR3)/IFN regulatory factor 7 (IRF7)-dependent type I IFN immunity to influenza virus. Of the 118 variants identified, 24 were validated as deleterious in ad hoc overexpression systems. The bona fide loss of function variants were found in 23 out of 659 patients with life-threatening COVID-19. In the 10 of 23 patients for whom samples were available, serum IFN-alpha levels were less than 1 pg/ml. In vitro, the defects identified led to reduced type I IFN production by plasmacytoid dendritic cells stimulated by SARS-CoV2 or impaired fibroblast-intrinsic type I IFN immunity to SARS-CoV2. The findings suggest that intact type IFN signaling can be critical to control SARS-CoV-2 infection. This is further supported by the companion manuscript by Paul Bastard and colleagues that found IFN autoantibodies in some patients with severe COVID-19 disease. Autoantibodies were detected against IFN-omega, the 13 types of IFN-alpha or both in 101 of 987 patients with severe COVID-19. No autoantibodies against type I IFN were detected in 663 people with asymptomatic or mild SARS-CoV-2 infection. Notably, 95 of the 101 patients with autoantibodies were men, and one of the women had X chromosome-linked incontinentia pigmenti, where cells activate only a single X chromosome. Plasma from eight patients with IFN auto-antibodies made cells permissive to infection with SARS-CoV-2 in the presence of IFN-alpha2. The autoantibodies were clinically silent before the patients developed COVID-19, suggesting that even the relatively modest amount of IFN induced by SARS-CoV-2 is important for protection against severe disease. Together, these studies highlight a central role for type I IFNs in protective immunity to SARS-CoV-2.