PERSONALIZED VACCINE BOOSTS T CELL RESPONSES
CT301 - A phase Ib study to evaluate RO7198457, an individualized Neoantigen Specific immunoTherapy (iNeST), in combination with atezolizumab in patients with locally advanced or metastatic solid tumors
Juanita S. Lopez, MB, BChir, MRCP, PhD (The Royal Marsden Hospital) discussed an early-phase trial of a personalized RNA-lipoplex vaccine for patients with solid cancers. The vaccine, known as RO7198457, was tested as a monotherapy and in combination with atezolizumab in dose escalation and expansion studies. The personalized vaccines were developed through NGS-based identification and prediction of immunogenic neoantigens, and patients were administered the RNA-based vaccine containing up to 20 neoantigens each.
Thirty patients were enrolled in the dose escalation cohort, while a total of 114 patients were included in the expansion group, including 72 checkpoint inhibitor-naïve patients. Optimal manufacturing of the vaccine took 5-6 weeks. Analysis of 63 patients to date indicated T cell responses to the predicted neoantigens in 73% of patients, with a median of 2.6 responses per patient by ELISPOT. No dose-limiting toxicities were observed, and those adverse events that did occur were mainly grade 1-2 and transient, the most common being infusion-related reactions. The dose escalation cohort included two responses to therapy: one complete response in a patient with CRC and one partial response in a TNBC tumor. In the expansion cohort, the overall response rate ranged from 4-30%, depending on the cancer type. The highest ORR was noted in patients with melanoma. Given the results in this study, further clinical trials are ongoing in selected cancer types.
COMBINATION TIGIT AND PD-L1 INHIBITION SHOWS EFFICACY IN PD-L1-POSITIVE TUMORS
CT302 - Phase Ia/Ib dose-escalation study of the anti-TIGIT antibody tiragolumab as a single agent and in combination with atezolizumab in patients with advanced solid tumors
A trial investigating an anti-TIGIT antibody was discussed by Johanna C. Bendell, MD (Sarah Cannon Research Institute). Phase 1A of the study tested tiragolumab monotherapy, while 1B combined tiragolumab with atezolizumab, with both arms studying the agents in patients with heavily pretreated solid tumors. Patients were allowed to cross over from arm A to B upon progression.
Monotherapy was tested in 24 patients in arm 1A and combination treatment was employed for 49 patients in 1B. At doses above 30 mg tiragolumab, no free TIGIT receptors could be found in general circulation. The recommended phase 2 dose was then 600 mg Q3W. In cohort 1A, no responses were noted, and 67% of patients experienced a treatment-related adverse event of any grade (4% grade 3+). Twelve patients from 1A crossed over to the combination treatment, but none of these patients experienced a response. In phase 1B, 59% of patients experienced a treatment-related adverse event and the ORR was 11%, with all responses occurring in PD-L1-positive tumors. An expansion cohort in PD-L1-positive NSCLC resulted in an ORR of 46% which included two complete responses. There was no evidence of pseudoprogression or delayed responses, nor any dose-limiting toxicities in this study, clearing the path for further investigations, including the recently-presented CITYSCAPE trial.
Read more about the biology of the TIGIT receptor in this recent article from the Journal for ImmunoTherapy of Cancer.
MOLECULAR CORRELATES WITH OUTCOMES IDENTIFIED IN HEAD AND NECK AND LUNG CANCERS
3384 - Characterization of the tumor microenvironment and liquid biopsy in head and neck and non-small cell lung cancer
Arutha Kulasinghe, PhD (Queensland University of Technology) and colleagues investigated the prognostic role of circulating tumor cells (CTCs) in head and neck (HNC) and lung cancers, as well as the immune profiles of the tumor microenvironments. CTCs were captured from 350 HNC patients and 150 patients with lung cancer and a range of enrichment platforms were tested to attempt short-term culture of the circulating cells. Digital spatial profiling was employed on tumor samples as well.
Many different platforms were tested for isolation of CTCs, including CellSearch, filtration, CD45 depletion, and various microfluidic chip technologies. Across these techniques, epitope-independent tools were able to capture the greatest proportion of CTCs. Mutations that were present in primary tumor tissues were also found in the CTCs, indicating that the CTCs accurately recapitulate the primary tissue. Overall, the presence of CTCs was associated with worse outcomes in HNC, including development of distant metastases and shorter progression-free survival. Additionally, PD-L1-positive CTCs associated with worse outcomes as well. A less pronounced trend was observed for the association of outcomes with CTC presence in NSCLC. Using spatial profiling, several immune markers were correlated with outcomes for head and neck cancers, including CD68, PD-1, CD44, STING, and others, while the CD8 levels did not associate with outcomes. For NSCLC, a tumor microarray was studied and markers including granzyme B, fibronectin and CD34 were found to be predictive of outcomes. These molecular analyses may help identify important biomarkers for cancer treatments in the future.
EXPLORATION OF MEDULLOBLASTOMA NEOANTIGENS IDENTIFIES IMMUNOGENIC PEPTIDES
1067 - Proteogenomic discovery of novel tumor proteins as neoantigens for personalized T cell immunotherapy in pediatric medulloblastoma
Samuel Rivero-Hinojosa, PhD (Children’s National Hospital) presented a study aimed at identifying neoantigens in medulloblastoma. As pediatric brain tumors are thought to have low mutation rates, identification of neoantigens for targeting cellular therapies is warranted.
This study employed a proteogenomic approach for identifying tumor-restricted peptides which were then used to separate T cells capable of responding to those peptides in a tumor-specific manner. Predicted novel proteins were generated using RNA-seq and WGS data from 46 medulloblastoma tumors, with proteins originating from somatic mutations, fusion proteins and novel isoforms. This database of predicted peptides was compared to mass spectrometry data from tumor samples, which enabled matched pairs of peptides between the two datasets to be identified. The main source of neoantigens in this study was found to be novel junctions, some of which were tested to determine their immunogenicity. Very few of the identified neoantigens were shared across patients. These predicted peptides were able to induce T cell responses, and these T cells were able to eradicate neoantigen-positive tumor cells in vitro. This study therefore points to the potential of cellular therapy in even those tumors with low mutation rates, like medulloblastoma.
A recent paper in the Journal for ImmunoTherapy of Cancer discussing biomarkers for glioblastoma can be read here.
IMMUNOGENIC NEOANTIGENS IDENTIFIED IN BREAST CANCERS
1068 - Tumor infiltrating lymphocytes (TIL) recognize a unique and diverse number of non-synonymous somatic mutations in patients with metastatic breast cancer
A study of tumor-infiltrating lymphocytes and tumor neoantigens from breast cancer was discussed by Nikolaos Zacharakis, PhD (National Cancer Institute). Using both WES and RNAseq, metastatic breast cancer lesions were analyzed for immunogenic non-synonymous mutations, which could serve as targets for TIL and adoptively transferred T cells. While mutations were being identified from tumor samples, harvested TIL were also expanded.
Every tumor lesion tested (n=39) expressed non-synonymous mutations (range: 6-1808, median: 102 mutations/patient), and TILs were successfully grown for all patients as well. The majority of patients (64%) had autologous TIL that recognized at least one of the mutated products, with 75% of the neoantigens recognized by CD4+ T cells. A median of four immunogenic peptides were identified per patient. None of the identified immunogenic peptides were shared across patients. TCR pairs were identified that recognized the immunogenic tumor mutations (99 TCR pairs, 50 peptides, 1-7 TCRs/peptide). While tumors like melanoma have benefited from TIL therapy due to their high mutational load, this study indicates that tumor-reactive, potentially therapeutic TIL are also present in other tumors, including breast cancer, and that immunogenic mutations are also present in these tumors.
MULTI-MODULAR CELL THERAPY APPROACH SHOWS PROMISE IN PROSTATE CANCER MODELS
1070 - AUTO7: Anti-PSMA humanized CAR T-cell with improved persistence and resistance to tumor microenvironment for metastatic castration resistant prostate cancer (mCRPC)
A multi-modular CAR T product was discussed by Marco Della Peruta, PhD (Autolus Therapeutics), designed to treat PSMA-positive prostate cancer. The cell product, termed AUTO7, contained components to target PSMA (CAR), induce TGF-1 pathway, induce proliferation through an IL-7 receptor, recruit and activate lymphocytes (through an IL-12 module), and a safety switch. Preclinical in vitro and in vivo studies were employed to test the efficacy and safety of the product.
The potency of the multi-modular product was found to be similar to a CAR alone control in cytotoxicity assays. The additional modules performed as expected: The addition of the TGFand PD-1 resistance to the CAR products enabled the T cells to withstand immunosuppression in vitro, and the constituently active IL-7 receptor also enabled cytokine-independent activity. The secretion of IL-12 by the CAR cells also was able to activate fresh PBMCs. Following this success in cellular assays, the CAR therapy was tested in PC3-PSMA-bearing NSG mice. The AUTO7 cells were able to completely eradicate tumor growth after intravenous administration, while the PSMA CAR-alone control could not. Minimal signs of toxicity were observed, indicating the therapy may be a safe and efficacious option for the treatment of PSMA-positive prostate cancer in the future.
HEXAVALENT CD40 AGONIST ENHANCES IMMUNE RESPONSES
1076 - Hexavalent HERA-CD40L induces a productive T cell-mediated anti-tumor immune response and shows superior activity in comparison to benchmark CD40 agonistic antibodies
Christian Gieffers, PhD (Apogenix AG) presented a preclinical investigation of HERA-CD40L, a hexavalent CD40 agonist comprised of two trivalent CD40L-receptor-binding domains and a silenced human IgG1 Fc domain. The investigators hypothesized that this molecular format would better mimic the natural ligand, enabling enhanced agonistic activity over traditional bivalent antibody formats.
HERA-CD40L was found to enhance the inflammatory activity of all CD40-expressing cells examined, including an increased and more rapid NFκB response in B cells and upregulation of activation markers in dendritic cells. Additionally, immature phagocytic macrophages were converted to mature, professional antigen-presenting cells after HERA-CD40L treatment in vitro, in contrast to traditional anti-CD40 antibody treatment, which resulted in polarization toward an M2 phenotype. Testing of a murine-specific variant of HERA-CD40L in tumor-bearing mice demonstrated single-agent activity of the molecule, modulating the tumor immune microenvironment and resulting in tumor growth indices of 29-43%, depending on the model. An increase in infiltration of antigen-specific T cells into tumor tissue was noted after the treatment, along with a decrease in M2 macrophages. These studies indicate that this agent may serve as an effective CD40 agonist, helping to modulate the immune response against tumors.
ROLE OF LYMPHANGIOGENESIS IN BREAST CANCERS ELUCIDATED
5695 - Induction of lymphangiogenesis in breast cancer enhances responsiveness to immunotherapies
The role of lymphangiogenesis in breast cancers was discussed by Peyman Hosseinchi, B.A.Sc (University of Chicago). While in some cases, lymphangiogenesis has been implicated in metastasis, increased tumor lymphatics may also boost the efficacy of immunotherapies. Therefore, the investigators aimed to investigate lymphangiogenesis in breast cancer with the goal of improving the efficacy of immunotherapies in this mostly resistant disease.
4T1 murine mammary cancer cells were engineered to overexpress VEGF-C, which induced increased tumor lymphangiogenesis in vivo. Compared to non-engineered 4T1 tumors, the VEGF-C-expressing tumors displayed higher levels of infiltrating CD4+ T cells and macrophages. This led to enhanced responsiveness to immunotherapy interventions including PD-1 and CTLA-4 inhibitors and a STING pathway agonist. However, when the activity of CD4+ cells or macrophages were inhibited, the efficacy of these immunotherapies was abrogated, indicating these cell populations are critical. Analysis of TCGA breast tumor data also indicated a correlation between VEGF-C expression and CD4+ T cell and macrophage infiltration, indicating these preclinical findings may translate into patients and potentially improve the efficacy of immunotherapies for breast cancer.
TRENDS IDENTIFIED RELATING ANTIBIOTIC USE AND COLORECTAL CANCER DEVELOPMENT
1055 - Antibiotic use and risk of colorectal cancer: A Swedish population-based registry study
The role of antibiotic use in the development of colorectal cancer (CRC) was explored by Sai San Moon Lu, MPH (Umeå University), given the known impact of the gut microbiota on this disease. This registry-based study incorporated cases and controls across the entire population of Sweden.
All primary CRC cases reported from 2010-2016 in the Swedish Colorectal Cancer Register were included in the study, ensuring at least five years’ follow up. Each CRC patient was matched to five cancer-free controls using age, sex and county of residence through the Register of the Total Population. Antibiotic use was extracted from various sources. Across the study population, which included over 40,000 CRC cases and over 200,000 controls, there was an association between antibiotic use and CRC risk, with an odds ratio of 1.11 for high antibiotic users relative to non-users. The trend was stronger still for proximal colon cancer (OR: 1.42) and for broad-spectrum beta lactams (OR: 1.50). There appeared to be a trend toward a reverse association between antibiotic use and rectal cancer, predominantly in women, wherein the highest users of antibiotics had the lowest risk of rectal cancer development. In addition, methenamine, an antibiotic without known impact on the gut microbiome, was not associated with CRC risk (OR: 0.92), indicating that modulation of the gut microbiota is critical in development of CRC.
You can read more about the role of the microbiome in cancer immunotherapy in this article from the Journal for ImmunoTherapy of Cancer.
ENGINEERED IL-18 VARIANT SHOWS NK CELL ACTIVATION AND ANTI-CANCER POTENTIAL
3424 - IL-18 immunotherapy is efficacious against checkpoint-immunotherapy refractory tumors by promoting the maturation of highly proliferative, polyfunctional NK cells
Ting Zhou, PhD (Yale University) discussed an immunotherapeutic strategy using a variant of IL-18. This “decoy-resistant” IL-18 (DR-18) was generated to activate the IL-18 receptor on NK cells while at the same time avoiding IL-18BP, an inhibitory decoy receptor that may be upregulated after immunotherapy. The group hypothesized that this agent would be able to stimulate NK cells to kill tumors that were lacking MHC-I expression.
DR-18 treatment was found to promote NK cell maturation, resulting in the production of effector molecules like TRAIL, IFN-B. The impact of this NK cell activation was tested in MC38 and YUMMER1.7 mouse models that were engineered to lack DR-18 treatment resulted in survival benefit and tumor regressions with a 30% cure rate. These results were further verified using B16F10 and RMA-S tumor models, which are known to be sensitive to NK cell-mediated cytotoxicity. When NK cells were ablated or IFN-γ was neutralized, the efficacy of DR-18 was removed as well. This study therefore represents a potential rational therapeutic option for tumors lacking MHC-I, a known resistance mechanism to other immunotherapies.
SMALL PORTION OF CERTAIN RARE CANCERS RESPOND TO COMBINATION IMMUNOTHERAPY
3417 - A phase II basket trial of dual anti-CTLA-4 and anti-PD-1 blockade in rare tumors (DART) SWOG S1609: The small bowel tumor cohort
3418 - A phase II basket trial of dual anti-CTLA-4 and anti-PD-1 blockade in rare tumors (DART) SWOG S1609: The salivary gland tumor cohort
Two cohorts from the SWOG S1609 study (dual anti-CTLA-4 & anti-PD-1 blockade in rare tumors, DART) were discussed in two presentations by Young Kwang Chae, MD, MPH, MBA (Northwestern University): the small bowel and the salivary gland tumor cohorts. This open-label phase 2 trial treated patients with a variety of rare tumors with ipilimumab 1 mg/kg Q6W and nivolumab 240 mg Q2W, with the primary endpoint of overall response rate.
The small bowel cohort included 23 patients with cancer of the duodenum, ileum, jejunum, or an unknown primary. The overall response rate was 9%, which included one complete response and one partial response. In this cohort, the median PFS was 2 months and the median OS was 6 months. Common adverse events included fatigue, diarrhea and dyspnea, with 70.8% of patients experiencing a treatment-related adverse event, including 41.7% of grade 3 or higher (including two deaths).
In the salivary gland cohort, patients were broken into two categories: those with adenoid cystic salivary gland tumors (n=26) or with other histologic subtypes (n=34), with the most common site of origin being the parotid gland. In the adenoid cystic group, the ORR was 4%, which contained one PR, and the median PFS and OS were 4.9 and 12 months, respectively. For patients with tumors of the remaining histologies, the ORR was 15% (five PRs), the median PFS was 4.9 months, and the median OS was not reached. Again, the most common toxicities included fatigue and diarrhea. Grade 3 or higher treatment-related adverse events occurred in 47.5% of patients, including one death.
Overall, these two presentations demonstrated limited success in treating rare tumors with combination CTLA-4 and PD-1 inhibition, indicating a small portion of patients may derive benefit from immunotherapy. Future biomarker investigations may help better refine the patient populations for these treatments.
Read about another study of immune checkpoint inhibition in rare cancers in the Journal for ImmunoTherapy of Cancer.
INVESTIGATING IMMUNE MICROENVIRONMENT OF PANCREATIC CANCER POINTS TO POTENTIAL THERAPEUTIC STRATEGY
3442 - Multimodal mapping of the immune landscape in human pancreatic cancer
Nina Steele, PhD (University of Michigan) discussed an in-depth study of the immune landscape of pancreatic cancer. Using single-cell RNAseq, the investigators analyzed pancreatic tumor cells from surgical specimens or biopsies as well as normal pancreas tissues.
Several signaling pathways were upregulated in the tested samples, including Hedgehog, NOTCH, and chemokine signaling. Differences were noted between T cells found in tumor tissues and healthy pancreas: increased expression of genes involved in T cell activation, exhaustion, and immune checkpoint pathways were seen in tumor-derived T cells. One of the most striking differences was in the level of TIGIT gene expression, with significantly increased expression in tumor T cells. TIGIT was also found to be nearly exclusively expressed on exhausted T cells rather than effector T cells, and TIGIT expression on tumor-infiltrating T cells and PBMCs correlated with one another. No difference in other checkpoints like PD-1 and LAG3 were noted between T cells in healthy and cancerous tissues. This investigation provided deep insight into the immune genetic landscape of the traditionally immunotherapy-resistant pancreatic tumor microenvironment and pointed to a possible therapeutic strategy incorporating TIGIT inhibition.
IMPACT OF REGULATORY T CELLS ON IFN-Γ-MEDIATED TUMOR REGRESSION INVESTIGATED
3444 - Intratumor regulatory T cells prevent IFN-gamma-dependent tumor vessel regression and can be selectively targeted by anti-CD25 near-infrared photoimmunotherapy.
The role of regulatory T cells in anti-tumor immunity was dissected by Yutaka Kurebayashi, MD, PhD (National Institutes of Health). To specifically deplete Tregs in the tumor microenvironment, this study employed a CD25-targeted photoimmunotherapy probe, which, after intravenous administration and near-infrared irradiation to the tumor, depleted only Tregs in the irradiated area.
When Tregs were removed from murine tumor tissues using the NIR-PIT strategy, the IFN-γ produced by T and NK cells mediated tumor regression. This study determined that the target of that IFN-γ was the endothelial cells of the tumor blood vessels. Upon Treg depletion, perivascular T and NK cells expressed IFN-γ for up to six hours, which caused vessels to regress, leading to intratumoral ischemia and tumor cell death. If IFN-γ was inhibited through an antibody or through genetic knock-out in mice, the therapeutic effects of the NIR-PIT treatment were abrogated. Treating mice with IL-15 in addition to NIR-PIT led to complete responses in MC38 and EO771 tumor models. Therefore, regulatory T cells in the tumor microenvironment clearly play a major role in anti-tumor immunity, and these interactions need to be further investigated.
A recent article in the Journal for ImmunoTherapy of Cancer discussed another way that regulatory T cells impact immune control of tumors.
COMBINATION IMMUNOTHERAPIES MAY NOT EXHIBIT SYNERGY
1047 - Combinatorial benefit without synergy in recent clinical trials of immune checkpoint inhibitors
Adam C. Palmer, PhD (University of North Carolina at Chapel Hill) presented an investigation of combinatorial immunotherapies, aimed at dissecting whether benefits from combination treatments were truly synergistic or due to the independent drug actions. The investigators utilized data from twelve randomized-controlled clinical trials of combination immunotherapy and other drug classes in several cancers, for which the activities of the individual monotherapies were also known.
The PFS curves for all treatment populations were digitized, and a PFS was computed for the null hypothesis of independent drug activity for combination treatments through random sampling of single-agent PFS curves. These computed combination and clinical PFS curves were then compared, and demonstrated that none of the analyzed trials of combination immunotherapy showed additive or synergistic effects beyond that of independent drug activity. PFS for combination treatments was largely predictable from the individual constituent drugs. Exploration of biomarker-selected populations, such as PD-L1-positive tumors, showed a similar trend. The study therefore hints that success from combination treatments may come from an increased chance of response with multiple therapies given at once, rather than the treatments interacting with one another.
Learn about the molecular impact of combination immunotherapies in this recent Journal for ImmunoTherapy of Cancer article.