Frontiers in Cancer Immunotherapy 2021

Immune checkpoint therapy has revolutionized cancer treatment, but there are many unanswered questions pertaining to mechanisms of response and resistance. In an attempt to address some of these questions, Dr. Sharma focuses her effort on a “reverse translation” process. She studies human immune responses to generate hypotheses related to mechanisms of tumor rejection, which she tests in appropriately designed pre-clinical models, and subsequently uses the new data to design novel clinical trials to improve outcomes for patients with cancer. For example, Dr. Sharma designed and conducted the first neoadjuvant (pre-surgical) trial with immune checkpoint therapy (anti-CTLA-4) in 2006. These studies in patients with localized bladder cancer led her to identify the ICOS/ICOSL pathway as a novel target for cancer immunotherapy strategies. This initial trial provided the first safety data for administration of immune checkpoint therapy in the neoadjuvant setting, which led to many other neoadjuvant studies. This trial was also the first demonstration of anti-tumor responses with immune checkpoint therapy for patients with bladder cancer, which led to many other clinical trials in patients with bladder cancer, including clinical trials with anti-PD-1 (nivolumab) that were led by Dr. Sharma and her colleagues, thereby enabling FDA-approval of nivolumab for patients with metastatic bladder cancer. In addition, Dr. Sharma demonstrated for the first time that human tumors express VISTA as an immunosuppressive pathway, which acts as a resistance mechanism to anti-CTLA-4 therapy. Clinical trials targeting VISTA are now ongoing. Dr. Sharma was also the first to demonstrate that anti-CTLA-4 plus inhibition of EZH2 can improve anti-tumor responses, which led her to design a new clinical trial with this combination. This clinical trial is currently accruing patients. More recently, Dr. Sharma demonstrated for the first time that the unique immunologic niche within a specific organ, such as bone, impacts response to treatment with immune checkpoint therapy. Dr. Sharma will discuss data from her research studies.

Treatment options are limited for advanced melanoma patients who progress after or do not respond to immune checkpoint inhibitors (ICI) and targeted therapies. Lifileucel is an adoptive cell therapy using TIL that has shown efficacy in patients with advanced melanoma who have progressed on/after anti-PD1 (Sarnaik 2020).  C-144-01 (NCT02360579) is a global Phase 2, multicenter study of lifileucel in advanced melanoma patients who have progressed on anti-PD1 therapy and BRAFi/MEKi, if BRAF V600 mutant. We report on Cohort 2 (N = 66) patients. Tumors were resected locally, shipped to central GMP facilities for manufacturing of TIL in a 22-day process. Therapy consisted of nonmyeloablative lymphodepletion (cyclophosphamide / fludarabine), a single lifileucel infusion, followed by up to 6 doses of IL-2. Objective response rate (ORR) was based on RECIST1.1. Data cutoff: 14Dec2020.  Baseline characteristics: 3.3 mean prior therapies (100% anti-PD1; 80% anti-CTLA4; 23% BRAFi/MEKi), high baseline tumor burden (106 mm mean target lesion SOD), 42% liver/brain lesions, 41% LDH > ULN. ORR by Investigator was 36.4% (3 CR, 21 PR). One patient converted from PR to CR at 24 mo. post-therapy. Median DOR was still not reached at median follow-up of 28 month of study follow up, and 11 patients (17%) had further SOD reduction since previous data cut of Apr2020.   Iovance is investigating lifileucel in earlier disease settings, in combination with ICI, and in other metastatic solid malignancies.

CD47 is an integral membrane protein that alters adaptive immunosurveillance when bound to the matricellular glycoprotein thrombospondin-1 (TSP1). We examined the impact of the TSP1/CD47 signaling axis on melanoma patient response to anti-PD1 therapy due to alterations in T cell activation, effector function and bioenergetics. An increase in T cells expressing CD47 and circulating levels of TSP1 post-anti-PD1 treatment was observed in non-responding patients compared to responding. Malignant melanoma patient biopsies also have increased CD47 expression compared to normal tissue. Therefore, TSP1/CD47 expression could serve as a marker to predict patient response to immune checkpoint blockade therapy. Additionally, TSP1 limited cytotoxic (CD8+) T cell effector function against melanoma cells; however, targeting CD47 allowed T cells to overcome this TSP1 interaction, sustaining activation and effector function. CD47 targeted naïve and activated CD8+ T cells have enhanced mitochondrial function and glycolysis compared to wild type, suggesting CD47 mediated metabolic reprogramming of T cells. A 50% decrease in tumor burden was observed by targeting CD47 in a syngeneic B16 melanoma model due to increased tumor oxygen saturation and granzyme B secreting CD8+ T cells. Therefore, the expression of TSP1/CD47 may serve as markers of immune checkpoint blockade response, and targeting this pathway may preserve T cell activation, effector function, and bioenergetics to reduce tumor burden.

Adoptive transfer of transplant donor-derived virus-specific T-cells is well recognized as a potentially curative therapy for EBV lymphomas, drug-refractory CMV infections and severe adenovirus, BK and JC infections. However, their application is limited by the time required for generating the T-cells and their cost. We have developed banks of EBV and CMV-specific T-cells, from healthy 3rd party donors, each characterized as to HLA type, virus-specificity, and HLA restriction. In our trials, such T-cells have induced durable remissions 68% of monoclonal Rituxan resistant EBV lymphomas developing in HCT recipients and 54% of solid organ transplant recipients who have failed Rituxan and CHOP chemotherapy. The method used to generate these T-cells also depletes alloreactive T-cells such that the risk GVHD or graft rejection is obviated. When engineered to express a tumor-reactive CAR, or TCRs mimic these EBVCTLs exhibit potent activity in animal models against hematologic malignancies. In early phase clinical trials, CAR-modified EBV-specific T-cells generated from patients or their HCT donors, have exhibited minimal toxicity and, in some cases, significant but short-term anti-tumor activity. At our center, early, ongoing trials of 3rd party-donor derived CAR modified EBVCTLs have shown particular promise in the treatment of refractory lymphomas, again with minimal risk of toxicities, cytokine release syndrome or GVHD. In this presentation, we will review the attributes and limitations of CAR-modified virus-specific T-cells and strategies being explored to enhance their immediate accessibility, broad applicability, effective anti-tumor activity and in vivo persistence.

We isolated mouse then human hematopoietic stem cells (HSCs). Transplantation of purified HSCs regenerates the blood and immune systems without graft vs. host disease, can induce transplant tolerance to tissues from the donor. In metastatic breast cancer patients post myeloablative chemotherapy, HSC purification allowed autologous transplantation of cancer-free grafts. The 15-year overall survival rate was ~33% in recipients of cancer-free HSCs.

To study the relationship between stem cells and cancer, we followed the preleukemic progression from hematopoietic stem cells (HSCs) to myelogenous leukemias, and found that the developing pre-cancerous HSC clones gradually accumulate mutations, with the last mutation giving rise to leukemia stem cells (LSC). The LSCs have acquired self-renewal and evaded programmed cell death and programmed cell removal.

A checkpoint inhibitor for innate immunity macrophages: Comparing LSC to HSC, we identified CD47 overexpression on LSC, then all cancers tested, and showed that it is used by cancer cells to evade macrophage phagocytosis by binding to SIRP alpha, its inhibitory receptor on macrophages. CD47 is the first target that is expressed on all human cancers tested. Humanized IgG4 anti-CD47 Abs are in clinical trials. anti-CD47 combination with Rituximab led to responses of aggressive lymphomas, relapsed and refractory to Rituxan and chemotherapy. The combined response rate was nearly 50%, 70-80% of these achieved complete remission.

CD47 is a widely expressed glycoprotein of the immunoglobulin superfamily that plays a critical role in self-recognition. Various solid and hematologic cancers exploit CD47 expression in order to evade immunological eradication, and its overexpression is clinically correlated with poor prognoses. CD47 interacts with signal regulatory protein-alpha (SIRPα) expressed on myeloid cells, transmitting inhibitory signals that regulate DC homeostasis, self-recognition, and macrophage-mediated phagocytosis. Given its essential role as a myeloid checkpoint for innate immunity and subsequent adaptive immunity, CD47-SIRPα axis has been explored as a new target for cancer immunotherapy. In this talk we will describe mechanistic insights into therapeutic modulation of the CD47 pathway and emerging clinical validation.

B-cell maturation antigen (BCMA) is a member of the tumor necrosis factor superfamily that has restricted expression on malignant and normal plasma cells making it a high priority target for novel multiple myeloma therapies. Numerous therapeutic approaches targeting BCMA are in clinical development, including chimeric antigen receptor (CAR) T cells, bispecific T cell engagers (TCE) and antibody drug conjugates (ADC). Idecabtagene vicleucel (ide-cel) is a BCMA CAR T cell therapy that has demonstrated a high frequency of tumor responses, including complete responses, in advanced myeloma patients with low rates of grade 3/4 cytokine release syndrome (CRS) and neurotoxicity (NT) (Raje N, New Engl J Med 2019, Munshi N, N Engl J Med 2021). FDA approval was granted for ide-cel in March 2021. Product and treatment related correlates of durable response and progression have been identified; BCMA antigen escape has been reported, but is rare. bb21217, a next-generation BCMA CAR T cell therapy contains the same CAR construct as ide-cel, but uses ex vivo culture in the presence of a PI3-kinase inhibitor to enrich the final product with memory-like T cells. NEX-T BCMA is another next generation BCMA-targeted CAR T cell therapy that has entered the clinic. T cell engaging bispecific antibodies, while not a cellular therapy, offer another approach to driving T cell mediated BCMA-targeted tumor killing. Taken together these data support the benefit of BCMA targeted, T cell activating immunotherapies for the treatment of advanced myeloma.

BCMA targeted CAR T cell therapy for relapsed/refractory myeloma induces unprecedented response rates in this patient population, however, many patients still relapse. We identified GPRC5D as a novel target for immunotherapy of myeloma and developed CAR T cells directed to this antigen. In mouse models we demonstrated that targeting GPRC5D can rescue BCMA-negative relapse. GPRC5D-targeted CAR T cells stemming from this work are currently under clinical investigation including for relapse post-BCMA CAR T cell therapy. In order to prevent antigen escape driven relapse, we then investigated multiple approaches to dual-targeted CAR T cell therapy.

Although NK cell-based therapies hold promise in the treatment of leukaemia, immunotherapy of solid cancers is still challenging. In particular, in solid tumors many factors in the hypoxic tumor microenvionment exist that counteract efficient NK cell anti-tumor activity. To obtain a comprehensive understanding of NK cells in solid tumors, we performed single cell RNA sequencing of RMAS-lymphoma infiltrated mouse NK cells compared to blood NK cells. Tumor-infiltrating NK cells were characterized by a signature reminiscent of dysfunctional NK cells. In particular, the transcription factor “Hypoxia Inducible Factor-1a; HIF-1a”, a crucial sensor of hypoxia in the tumor microenvironment, was upregulated in tumor NK versus blood NK cells. Conditional deletion of Hif1a in NK cells resulted in reduced tumor growth, elevated expression of activation markers, effector molecules, and an enriched NF-κB pathway in tumor-infiltrating NK cells. Apart from cytotoxic NK cells, ILCs comprise helper cytokine-producing Innate Lymphoid Cell subsets (ILC 1-3) that are present in many solid tumors. Their role in cancer is still controversial and will be further discussed in the presentation.

Immunotherapies targeting the interaction of programmed death 1 (PD-1) with its ligands, PD-L1 and PD-L2, have revolutionized modern oncology. The PD-1 pathway is a key mediator of local immunosuppression in the tumor microenvironment. In advanced treatment-refractory cancers, drugs blocking this pathway can mediate tumor regression. While anti–PD-(L)1 has demonstrated a broad activity profile and is regarded as a “common denominator” for cancer therapy, many advanced inoperable tumors demonstrate de novo or acquired treatment resistance. One potential solution is to use anti–PD-(L)1 in the neoadjuvant (pre-surgical) setting, when the tumor is potentially “resectable for cure”, to eliminate micrometastatic deposits that could otherwise cause post-surgical relapse. The development of neoadjuvant immunotherapies is based upon immunologic mechanisms and clinical considerations. This treatment approach offers novel clinical endpoints, such as pathologic response assessments. Furthermore, it enables in-depth studies of complete surgical resection specimens to discover biomarkers and reveal new insights into the mechanisms of action of the PD-1 pathway. Hundreds of clinical trials are now underway testing neoadjuvant immunotherapy across a wide variety of cancers.

A breakthrough in the field of regenerative medicine and cellular immunotherapeutics came with the introduction of induced pluripotent stem cells (iPSCs).  iPSCs offer an unlimited supply of autologous cells that could be used without the risk of immune rejection. However, problems arose with autologous transplantation of iPSCs and their differentiated end-products, including high cost and excessive manufacturing time, making them clinically impractical to use. Therefore, the solution is to use donor iPSC cell lines to produce allogeneic “off-the-shelf” tissue matching cell products. Although this has reduced the manufacturing time, it is still a relatively expensive exercise that clinically demands immunosuppression. Hypothetically, a naturally occurring PSC found in abundance would have low cost, less manufacturing time, and used autologously, would eliminate immunosuppression protocols. We have discovered, using immunohistochemical analysis, a population of cells in human peripheral blood that are pluripotent, easy to isolate, and abundant. This discovered population of cells range between 1 to 5 million cells per ml of plasma, are relatively small in diameter (<5um), express all four Yamanaka factors (OCT4, SOX2, MYC, KLF4), stain positive for the Kyoto Probe (KP-1), and express Nanog, CD133, CXCR4, SSEA3 and SSEA4. These easily obtainable, naturally occurring PSCs from peripheral blood may have positive implications in generation of tumor-targeting T/NK cells.

Hepatocellular carcinoma (HCC) represents a major health concern, causing over 700,000 deaths annually. Recently, immunotherapies have revolutionized treatment of advanced HCC but, unfortunately, <30% of patients are sensitive, indicating underlying resistance to these immunotherapies. To understand mechanisms of HCC immune evasion, we performed transcriptomic analysis on a collection of immune-escaped murine tumors. We identified an enrichment signature of Notch signaling in immune inflamed tumors that was associated with a specific deficit of CD8+ T cells – a result validated in HCC patient samples. To test the role of Notch in immune escape, we generated a novel mouse model of HCC based on MYC overexpression, activated Notch (MYC;NICD1), and customized expression of tumor antigens (tAg). While expression of tAG in MYC;p53-/- mice led to enhanced survival and immune surveillance compared to MYC;p53-/- mice without tAg, this survival advantage conferred by tAg expression was abrogated in MYC;NICD1 mice, demonstrating that Notch activation drives immune evasion in HCC. Mechanistically, Notch activation led to a reduced number of tumor antigen-specific CD8+ T cells compared to the MYC;p53-/- mice. We are exploring the effects of Notch activation on anti-PD1 response in mice and HCC patients. Together, Notch activation promotes immune escape in HCC and may lead to resistance to anti-PD1 therapy, implicating Notch pathway as a putative biomarker for HCC patient stratification.

Evidence supports complex subclonal relationships in solid tumours, manifested as intratumour heterogeneity. Parallel evolution of subclones, with distinct somatic events occurring in the same gene, signal transduction pathway or protein complex, suggests constraints to tumour evolution that might be therapeutically exploitable. Data from TRACERx, a longitudinal lung cancer evolution study will be presented. Drivers of tumour heterogeneity change during the disease course and contribute to the temporally distinct origins of lung cancer driver events. APOBEC driven mutagenesis appears to be enriched in subclones in multiple tumour types. Oncogene, tumour suppressor gene and drug induced DNA replication stress are found to drive APOBEC mutagenesis. Phylogenetic tracking detects minimal residual disease and clonal evolution of disease from primary to metastatic sites, presenting opportunities for drug development.

On-going chromosomal instability, manifested as Mirrored Subclonal Allelic Imbalance (MSAI) is found to be a major driver of intratumour heterogeneity across cancer types, contributing to parallel evolution and selection. Subclonal driver events, evidence of ongoing selection within subclones, combined with genome instability driving cell-to-cell variation is likely to limit the efficacy of targeted monotherapies, suggesting a need for new approaches to drug development and integration of cancer immunotherapeutic approaches. Multiple adaptive mechanisms to neo-antigen evolution have been found in TRACERx highlighting cancer chromosomal instability driving immune evasion and HLA loss and loss of clonal neo-antigens as well as epigenetic repression of neo-antigens. The clonal neo-antigenic architecture may act as a tumour vulnerability to mitigate resistance and treatment failure.

Allogeneic transplantation using grafts from matched unrelated donors are the most commonly performed transplant procedures currently. Survival is similar to transplants using a graft from an HLA-identical sibling donor. Although HLA is a major factor in selecting an unrelated donor, several other genetic and non-genetic factors are also of importance. Despite the high number of volunteer unrelated donors listed on international registries not all patients, especially those from racial and ethnic minorities, will have an available donor. Novel strategies are being tested to improve the outcome of transplants using mismatched unrelated donors.

Unmanipulated double unit cord blood transplantation (CBT) can be highly efficacious in adults with high-risk hematologic malignancies with comparable progression-free survival (PFS) to that of adult donor allografts in many series. However, as CBT can be associated with delayed hematopoietic recovery, a major focus in the CBT field has been ex vivo expansion. While expansion may be appealing, it has the potential to be logistically challenging. Moreover, the re-frozen/re-thawed T-cell enriched CD34-negative add-back may result in a loss of viable T-cells which could potentially compromise the immune properties of the graft. As an alternative we have investigated unmanipulated double unit CBT in adults with high risk hematologic malignancies using optimized unit selection and transplant practices including a novel intermediate intensity conditioning regimen. This therapy has been associated with very promising outcomes in middle aged adults over 50% of whom have had part or full non-European origins and diagnoses (acute leukemia or other hematologic malignancy) thought to be otherwise incurable. Specifically, both transplant-related mortality and relapse were very low resulting in a 3-year PFS probability of over 75%. Notably, neither the presence of minimal residual disease (MRD) pre-transplant nor engrafting with a highly HLA-mismatched (3-4/8) unit adversely impacted survival supporting the unique immunobiology associated with these transplants. Very promising PFS was observed in AML patients with a high relapse risk according to their diagnosis ELN classification and those with FLT-3 ITD mutations. Transplantation of these cryopreserved, HLA-mismatched, unmanipulated CB grafts after intermediate intensity conditioning should permit more widespread adoption of CBT. It can result in a major extension of allograft access to racial and ethnic minority patients, and those with high risk disease who should not wait to secure an available adult donor or could benefit from the transplantation of a T-replete product. Thus, our findings support further investigation of intermediate intensity double unit CBT in a multi-center setting. Furthermore, we are now using this platform to develop next generation CBT transplants to even further improve transplant outcomes.