A Phase 3 Randomized Clinical Trial of Chemotherapy With or Without Algenpantucel-L (HyperAcute-Pancreas) Immunotherapy in Subjects With Borderline Resectable or Locally Advanced Unresectable Pancreatic Cancer

OBJECTIVES

To compare the efficacy and safety of algenpantucel-L [HyperAcute-Pancreas algenpantucel-L (HAPa); IND# 12311] immunotherapy combined with standard of care (SOC) chemotherapy and chemoradiation to SOC chemotherapy and chemoradiation therapy alone in patients with borderline resectable or locally advanced pancreatic ductal adenocarcinoma (PDAC).

SUMMARY BACKGROUND DATA

To date, immunotherapy has not been shown to benefit patients with borderline resectable or locally advanced unresectable PDAC. HAPa is a cancer vaccine consisting of allogeneic pancreatic cancer cells engineered to express the murine α(1,3)GT gene.

METHODS

A multicenter, phase 3, open label, randomized (1:1) trial of patients with borderline resectable or locally advanced unresectable PDAC. Patients received neoadjuvant SOC chemotherapy (FOLFIRINOX or gemcitabine/nab-paclitaxel) followed by chemoradiation (standard group) or the same standard neoadjuvant regimen combined with HAPa immunotherapy (experimental group). The primary outcome was overall survival.

RESULTS

Between May 2013 and December 2015, 303 patients were randomized from 32 sites. Median (interquartile range) overall survival was 14.9 (12.2-17.8) months in the standard group (N = 158) and 14.3 (12.6-16.3) months in the experimental group (N = 145) [hazard ratio (HR) 1.02, 95% confidence intervals 0.66-1.58; P = 0.98]. Median progression-free survival was 13.4 months in the standard group and 12.4 months in the experimental group (HR 1.33, 95% confidence intervals 0.72-1.78; P = 0.59). Grade 3 or higher adverse events occurred in 105 of 140 patients (75%) in the standard group and in 115 of 142 patients (81%) in the experimental group (P > 0.05).

CONCLUSIONS

Algenpantucel-L immunotherapy did not improve survival in patients with borderline resectable or locally advanced unresectable PDAC receiving SOC neoadjuvant chemotherapy and chemoradiation.

TRIAL REGISTRATION

ClinicalTrials.gov Identifier: NCT01836432.

Effect of Taxane Chemotherapy With or Without Indoximod in Metastatic Breast Cancer: A Randomized Clinical Trial

Importance

Indoleamine 2,3-dioxygenase 1 (IDO1) causes tumor immune suppression. The IDO1 pathway inhibitor indoximod combined with a taxane in patients with ERBB2-negative metastatic breast cancer was tested in a prospective clinical trial.

Objective

To assess clinical outcomes in patients with ERBB2-negative metastatic breast cancer treated with indoximod plus a taxane.

Design, Setting, and Participants

This phase 2 double-blinded randomized 1:1 placebo-controlled clinical trial enrolled patients at multiple international centers from August 26, 2013, to January 25, 2016. Eligibility criteria included ERBB2-negative metastatic breast cancer, ability to receive taxane therapy, good performance status, normal organ function, no previous immunotherapy use, and no autoimmune disease. The study was discontinued in June 2017 because of lack of efficacy. Data analysis was performed from February 2019 to April 2020.

Interventions

A taxane (paclitaxel [80 mg/m2] weekly 3 weeks on, 1 week off, or docetaxel [75 mg/m2] every 3 weeks) plus placebo or indoximod (1200 mg) orally twice daily as first-line treatment.

Main Outcomes and Measures

The primary end point was progression-free survival (PFS); secondary end points were median overall survival, objective response rate, and toxic effects. A sample size of 154 patients would detect a hazard ratio of 0.64 with 1-sided α = .1 and β = .2 after 95 events. Archival tumor tissue was stained with immunohistochemistry for IDO1 expression as an exploratory analysis.

Results

Of 209 patients enrolled, 169 were randomized and 164 were treated (85 in the indoximod arm; 79 in the placebo arm). The median (range) age was 58 (29-85) years; 166 (98.2%) were female, and 135 (79.9%) were White. The objective response rate was 40% and 37%, respectively (indoximod vs placebo) (P = .74). The median (range) follow-up time was 17.4 (0.1-39.4) months. The median PFS was 6.8 months (95% CI, 4.8-8.9) in the indoximod arm and 9.5 months (95% CI, 7.8-11.2) in the placebo arm (hazard ratio, 1.2; 95% CI, 0.8-1.8). Differences between the experimental and placebo arms in median PFS (6.8 vs 9.5 months) and overall survival (19.5 vs 20.6 months) were not statistically significant. Grade 3 or greater treatment-emergent adverse events occurred in 60% of patients in both arms.

Conclusions and Relevance

This randomized clinical trial found that, among patients with ERBB2-negative metastatic breast cancer, addition of indoximod to a taxane did not improve PFS compared with a taxane alone.

Trial Registration

ClinicalTrials.gov Identifier: NCT01792050.

Discovery of indoximod prodrugs and characterization of clinical candidate NLG802

A series of different prodrugs of indoximod, including estesrs and peptide amides were synthesized with the aim of improving its oral bioavailability in humans. The pharmacokinetics of prodrugs that were stable in buffers, plasma and simulated gastric and intestinal fluids was first assessed in rats after oral dosing in solution or in capsule formulation. Two prodrugs that produced the highest exposure to indoximod in rats were further tested in Cynomolgus monkeys, a species in which indoximod has oral bioavailability of 6-10% and an equivalent dose-dependent exposure profile as humans. NLG802 was selected as the clinical development candidate after increasing oral bioavailability (>5-fold), C_max (6.1-3.6 fold) and AUC (2.9-5.2 fold) in monkeys, compared to equivalent molar oral doses of indoximod. NLG802 is extensively absorbed and rapidly metabolized to indoximod in all species tested and shows a safe toxicological profile at the anticipated therapeutic doses. NLG802 markedly enhanced the anti-tumor responses of tumor-specific pmel-1 T cells in a melanoma tumor model. In conclusion, NLG802 is a prodrug of indoximod expected to increase clinical drug exposure to indoximod above the current achievable levels, thus increasing the possibility of therapeutic effects in a larger fraction of the target patient population.

A Phase I Study of Alpha-1,3-Galactosyltransferase-Expressing Allogeneic Renal Cell Carcinoma Immunotherapy in Patients with Refractory Metastatic Renal Cell Carcinoma

LESSONS LEARNED

HyperAcute Renal immunotherapy was well tolerated and demonstrated antitumor activity in patients requiring salvage-line treatment for metastatic renal cell carcinoma (mRCC). HyperAcute Renal immunotherapy was safely administered with concomitant salvage-line treatments for mRCC, and it may be a candidate for inclusion in novel combinations for salvage treatment of mRCC because of its unique mechanism of action.

BACKGROUND

HyperAcute Renal (HAR) immunotherapy exploits a naturally occurring barrier to xenotransplantation and zoonotic infections in humans to immunize patients against metastatic renal cell carcinoma (mRCC) cells. HAR consists of two allogeneic renal cancer cell lines genetically modified to express α(1,3)Gal, to which humans have an inherent pre-existing immunity.

METHODS

Patients with refractory mRCC were eligible for this phase I dose-escalation trial. Concomitant treatment was permitted after the initial 2 months of HAR monotherapy. HAR was injected intradermally weekly for 4 weeks then biweekly for 20 weeks, totaling 14 immunizations. The primary endpoint was safety and determination of a maximum tolerated dose (MTD).

RESULTS

Among 18 patients enrolled, two grade 3 adverse events (AEs) were attributed to HAR, lymphopenia and injection site reaction, and no grade 4/5 AEs occurred. The recommended phase II dose (RP2D) was 300 million cells. One patient had a partial response and eight patients had stable disease, for a disease control rate of 50% (9/18). Median overall survival with low-dose HAR was 14.2 months and was 25.3 months with high-dose HAR.

CONCLUSION

In pretreated mRCC, HAR immunotherapy was well tolerated and demonstrated antitumor activity. HAR immunotherapy may be a candidate for inclusion in novel combinations for salvage treatment of mRCC.

Novel suspension cell-based vaccine production systems for Rift Valley fever virus-like particles

Rift Valley fever virus (RVFV) is an arthropod-borne pathogen that often results in severe morbidity and mortality in both humans and livestock. As its geographic range continues to expand, it presents a real threat to naïve populations around the world by accidental introduction (e.g., the result of increased travel) or intentional release (e.g., a bioterror event). While there is a clear need for a safe and efficacious vaccine against this emerging and re-emerging pathogen, no FDA-approved vaccine is currently available. This need was addressed by the establishment of novel mammalian and insect suspension cell line systems for the efficient production of RVF virus-like particle (VLP)-based vaccine candidates. A direct comparison of the production of RVF VLPs in these systems was performed. Optimization and characterization resulted in a production platform suitable for scale-up. Furthermore, RVF VLP-based vaccines were tested in a lethal challenge model and showed full protection, demonstrating that RVF VLPs present promising RVFV vaccine candidates.

Abstract 2423: Phase I/II study of antitumor vaccination using lung cancer cells expressing murine α(1,3)galactosyltransferase (αGT) in non-small cell lung cancer (NSCLC)

Background: Alpha(1,3)-linked galactose (αGal) epitopes are immunogenic and mediate rejection of non-primate xenografts. Vaccination of αGal knockout mice with tumor expressing αGT stimulated immunity against αGal-negative tumor and induced rejection of tumor. We studied the safety and antitumor activity of a vaccine using allogeneic human NSCLC cells engineered to express murine αGT gene in patients with NSCLC. The vaccine is composed of equal numbers of three αGT transduced human NSCLC cell lines. Methods: Patients with metastatic or recurrent NSCLC, age ≥18, ECOG PS ≤2, ≤2 prior systemic therapies, adequate hematological, hepatic and renal function, and informed consent were eligible. In phase I cohorts of patients received intradermal injections totaling 3, 10, 30, or 100 × 10E+6 vaccine cells every 4-weeks x 4; or 500 × 10E+6 vaccine cells followed by 300 × 10E+6 vaccine cells every 2-weeks × 7. In Phase II, eligible patients received injections totaling 300 × 10E+6 vaccine cells every 2 weeks for 8 doses. Adverse events were assessed using CTC v3.0 and response was determined using modified RECIST criteria. Serum anti-αGal antibody titers were monitored and ELISPOT assays for induction of interferonγ and interleukin-5 by PBMC were examined pre- and post vaccination. Vaccine site biopsies were performed. Results: 17 patients were treated in phase I. No vaccine-related serious adverse events (AE) were observed. Related AE's were ≤CTC grade 2 and included injection site pain/discomfort, local hyperpigmentation, skin reactions, local urticaria, arthralgias/myalgias, dyspnea, fatigue, herpes zoster, hypertension, lymphopenia and mild serum transaminitis. Six patients experienced stable disease for ≥4.0 months (median 8.0, range, 6.0-63.1 months). In phase II, 23 patients treated, 16 patients received one and five patients received two prior treatment regimens and two patients had no prior treatment (one of those had surgery only). Seven patients demonstrated SD ≥4.0 months. Median survival for the entire group was 11.0 months (range, 2.1-35.7 months) and 23.7 months (range, 11.0-35.7 months) for patients that achieved SD. Vaccine sites demonstrated infiltration of lymphocytes, granulocytes and eosinophils. Anti-αGal titers increased in all patients, and release of interferonγ and interleukin-5 during co-culture with vaccine parental and an unrelated NSCLC cell line was detected in some patients. Increases in anti-CEA antibodies were also detected. Conclusions: Antitumor vaccination using genetically altered human lung cancer cells expressing αGT is safe and feasible. A number of patients had prolonged SD and the median survival of vaccinated patients compared favorably to that reported in patients receiving 2nd line chemotherapy for relapsed or progressive advanced NSCLC.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 2423.

Allogeneic melanoma vaccine expressing alphaGal epitopes induce anti-tumor immunity to autologous melanoma antigens without signs of toxicity (48.7)

The hyperacute rejection of a xenotransplant is a potent mechanism of tissue destruction mediated by complement-fixing natural Ab that recognize alphaGal (aGal) epitopes. These epitopes are not present on human cells. Humans develop high titers of naturally acquired Ab against these epitopes. We hypothesized that this mechanism could be used to generate a potent anti-cancer vaccine. In this study we tested this hypothesis by using an allogeneic whole cell cancer vaccine to treat murine melanoma. We used the aGal KO mice (aGT H-2bb) and the aGal negative melanoma cell lines B16 and S91M3. We genetically engineered S91M3 (H-2dd) cells to produce the aGal(+)S91M3 cells. Mice with pre-existing subcutaneous and pulmonary B16 (H-2bb) tumors were efficiently treated with aGal(+)S91M3 melanoma vaccines. Control S91M3 vaccines had no impact on tumor growth. Vaccination with aGal(+)S91M3 induced cytotoxic CD8+ T cells recognizing the “autologous” aGal(-)B16 tumors measured by IFN-gamma release and by adoptive transfer experiments. The presence of allo-antigens did not dominate the induction of immunity to “cryptic” tumor antigens tested in the pre-existing melanoma metastasis model. The effective anti-tumor immunity was not associated with autoimmunity or other signs of toxicity in long-term toxicology studies for breast, lung and melanoma allogeneic vaccines. This study establishes for the first time the safety and efficacy data of allogeneic aGal(+) whole cell vaccines and constituted the basis for the initiation of human clinical trials to treat human cancers currently underway.

A phase I study of antitumor vaccination using tumor cells genetically modified to express alpha(1,3)galactosyltransferase (αGT) in patients with refractory or recurrent non-small cell lung cancer (NSCLC)

12503

Background: Despite new treatments, lung cancer remains the leading cause of cancer death. We examined the safety and activity of antitumor vaccination using genetically altered human non-small cell lung cancer cells (HAL; HyperAcute Lung Cancer Vaccine) engineered to express xenotransplantation antigens through retroviral transfer of the murine αGT gene in patients (Pts.) with NSCLC. Methods: A single institution Phase I trial. Eligibility: Stage IV, recurrent or progressive NSCLC, Age ≥18, ECOG PS ≤2, prior chemotherapies ≤2, AGC ≥1,500/μL, Plts. ≥100,000/μL, adequate hepatic/renal function and informed consent. Cohorts of Pts. received intrademal injections of 3 × 106, 10 × 106, 30 × 106, or 100 × 106 HAL vaccine cells every 4-weeks × 4, or 500 × 106 HAL cells followed by 300 × 106 HAL cells every 2-weeks × 7. Toxicity was assessed using CTCv3.0 and response by RECIST criteria. Immunological responses included anti-αGal antibody titers, interferon-γ ELISPOT and vaccine site skin biopsies. Results: Seventeen Pts., 9 men and 8 women, median age 57 years (range, 34–85), median number of prior chemotherapies 1 (range, 1–2) were vaccinated. There were no Grade 3/4 adverse events attributable to the study vaccine. Adverse events (≤grade 2) attributable to vaccination include injection site urticaria, pain/discomfort, local skin reaction, fatigue, herpes zoster, arthralgias/myalgias and hypertension. Six Pts. had stable disease ≥16 weeks duration (range, 16–85+ weeks), 9 Pts. progressed and 2 Pts. were not evaluable for response. Skin biopsies 48-hrs after vaccination demonstrated vaccine cells in the dermis with acute infiltration of lymphocytes, granulocytes and eosinophils. Serum anti-αGal titers increased 10–14 fold with vaccination. Conclusions: Xenogeneic antitumor vaccination with genetically altered allogeneic human lung cancer cells expressing αGT is safe and feasible. A phase II trial is planned.

No significant financial relationships to disclose.

Rat sodium iodide symporter allows using lower dose of 131I for cancer therapy

Efficient gene delivery is a critical obstacle for gene therapy that must be overcome. Until current limits of gene delivery technology are solved, identification of systems with bystander effects is highly desirable. As an anticancer agent, radioactive iodine (131)I has minimal toxicity. The physical characteristics of (131)I decay allow radiation penetration within a local area causing bystander killing of adjacent cells. Accumulation of (131)I mediated by the sodium iodide symporter (NIS) provides a highly effective treatment for well-differentiated thyroid carcinoma. Other types of cancer could also be treated by NIS-mediated concentration of lethal (131)I radiation in tumor cells. Our group and others previously reported that a significant antitumor effect in mice was achieved after adenoviral delivery of rat or human NIS gene following administration of 3 mCi of (131)I. We have also demonstrated 5-6-fold greater uptake of (125)I by rat NIS over human NIS in human cancer cells. Recently, we reported the capability of the rat NIS and (131)I to effectively induce growth arrest of relatively large tumors (approximately 800 mm(3)) in an animal model. In the present work tumor growth inhibition was achieved using adenoviral delivery of the rat NIS gene and 1 mCi of (131)I (one-third of the dose used in earlier reports). We also demonstrated that a higher concentration of (123)I was accumulated in the NIS-expressing tumors than in the thyroid 20 min after radioiodine administration. The highest intratumoral radioiodine concentration was observed along the needle track; however, the rat NIS-(131)I effectively induced growth arrest of tumor xenografts in mice through its radiological bystander effect. Importantly, the rat NIS allowed reducing the injected radioiodine dose by 70% with the same antitumor efficacy in pre-established tumors. These results suggest that the rat NIS gene may be advantageous compared to the human gene in its ability to enhance intratumoral (131)I uptake.

Effective treatment of preexisting melanoma with whole cell vaccines expressing alpha(1,3)-galactosyl epitopes

The hyperacute immune response in humans is a potent mechanism of xenograft rejection mediated by complement-fixing natural antibodies recognizing alpha(1,3)-galactosyl epitopes (alphaGal) not present on human cells. We exploited this immune mechanism to create a whole cell cancer vaccine to treat melanoma tumors. B16 melanoma vaccines genetically engineered to express alphaGal epitopes (B16alphaGal) effectively treated preexisting s.c. and pulmonary alphaGal-negative melanoma (B16Null) tumors in the alpha(1,3)-galactosyltransferase knockout mouse model. T cells from mice vaccinated with B16alphaGal recognized B16Null melanoma cells measured by detection of intracellular tumor necrosis factor-alpha. We showed successful adoptive transfer of immunity to recipient mice bearing lung melanoma metastasis. Mice receiving lymphocytes from donors previously immunized with B16alphaGal had reduced pulmonary metastases. The transfer of lymphocytes from mice vaccinated with control vaccine had no effect in the pulmonary metastasis burden. This study unequivocally establishes for the first time efficacy in the treatment of preexisting melanoma tumors using whole cell vaccines expressing alphaGal epitopes. Vaccination with B16alphagal induced strong long-lasting cell-mediated antitumor immunity extended to B16Null. These data formed the basis for the testing of this therapeutic strategy in human clinical trials currently under way.

Adeno-retroviral chimeric viruses as in vivo transducing agents

Several hybrid viral gene transfer systems have been described that exploit the favorable features of the two parent viral species. We have developed a hybrid adeno-retroviral vector system to generate a retroviral vector in situ. The system consists of adenoviruses encoding MoMLV gag.pol (Axtet.gag.pol), the VSV-G viral envelope (Axtet.VSV-G), the retroviral vector LXSN expressing the neomycin phosphotransferase gene (AV-LXSN) and a transcriptional regulator to control expression of gag.pol and envelope (AV-rtTA). In vitro, biologically active retroviral vector preparations were generated following adeno-retroviral transduction of 9L rat glioma cells. In vivo the transcomplementing adeno-retroviruses were co-administered intratumorally into subcutaneous 9L glioma tumors in rats and human A375 melanoma xenografts in nude mice. In the 9L rat model, G418 cell cultures were only obtained when 9L cells were harvested from tumors injected with all four transcomplementing adeno-retroviruses. Molecular analysis of DNA extracted from 9L G418 populations derived both in vitro and in vivo showed appropriate integration of the LXSN proviral sequence. Tumor cells were harvested 1, 3 and 4 weeks after adeno-retrovirus administration to the human A375 xenografts. The percentage of G418 colonies recovered from tumors transduced with all of the transcomplementing adeno-retroviruses increased with time, whereas no increase was observed in tumors transduced with AV-LXSN alone. DNA extracted from G418 A375 cell populations showed the presence of integrated proviral sequences only in animals that received the full complement of adeno-retroviruses. These results demonstrate that adenoviral vectors expressing transcomplementing genes for retroviral proteins and retroviral vector RNAs can be used for in situ transduction of target cells.

Enhancement of the herpes simplex virus thymidine kinase/ganciclovir bystander effect and its antitumor efficacy in vivo by pharmacologic manipulation of gap junctions

Apigenin, a flavinoid, and lovastatin, an HMG-CoA reductase inhibitor, upregulated gap junction (GJ) function and dye transfer in tumors expressing GJ and were inactive in the GJ-negative tumor line N2a. N2a cells transfected with the connexin 43 gene showed restored cell-to-cell dye transfer, which could then be improved nearly fourfold by addition of apigenin. To test the drugs in HSV thymidine kinase/ganciclovir (HSV-tk/GCV) tumor killing, mixtures of 90% wild-type (WT) with 10% HSV-tk gene-modified MCA38 adenocarcinoma cells were exposed in vitro to GCV +/- apigenin or lovastatin. A significant bystander effect (BSE) was seen following GCV treatment alone, while neither apigenin or lovastatin alone had any effect on the recovery of viable tumor colonies. However, GCV-treated cultures also exposed to apigenin or lovastatin showed an increased BSE and reduced tumor cell recovery. Thirty percent of mice bearing tumors from the same mixture of 90% WT and 10% HSV-tk MCA38 cells treated with GCV alone became tumor free. Tumor-bearing mice given only two or three injections of lovastatin or apigenin during GCV treatment had a doubling of the antitumor response rate, with 60-70% of the mice achieving complete remission. These results support the hypothesis that the transfer of phosphorylated GCV from HSV-tk gene-expressing cells to neighboring WT tumor cells is a major component of the BSE and that pharmacological manipulation of GJ function with lovastatin or apigenin can result in striking improvement in the antitumor response in mice with tumors modified to contain as few as 10% HSV-tk cells.