LOAd703, an oncolytic virus-based immunostimulatory gene therapy, combined with chemotherapy for unresectable or metastatic pancreatic cancer (LOKON001): results from arm 1 of a non-randomised, single-centre, phase 1/2 study

Abstract

Background: Pancreatic ductal adenocarcinoma is characterised by low immunogenicity and an immunosuppressive tumour microenvironment. LOAd703, an oncolytic adenovirus with transgenes encoding TMZ-CD40L and 4-1BBL, lyses cancer cells selectively, activates cytotoxic T cells, and induces tumour regression in preclinical models. The aim of this study was to evaluate the safety and feasibility of combining LOAd703 with chemotherapy for advanced pancreatic ductal adenocarcinoma.

Methods: LOKON001 was a non-randomised, phase 1/2 study conducted at the Dan L Duncan Comprehensive Cancer Center, Baylor College of Medicine, Houston, TX, USA, and consisted of two arms conducted sequentially; the results of arm 1 are presented here. In arm 1, patients 18 years or older with previously treated or treatment-naive unresectable or metastatic pancreatic ductal adenocarcinoma were treated with standard 28-day cycles of intravenous nab-paclitaxel 125 mg/m² plus gemcitabine 1000 mg/m² (up to 12 cycles) and intratumoural injections of LOAd703 every 2 weeks. Patients were assigned using Bayesian optimal interval design to receive 500 μL of LOAd703 at 5 × 10¹⁰ (dose 1), 1 × 10¹¹ (dose 2), or 5 × 10¹¹ (dose 3) viral particles per injection, injected endoscopically or percutaneously into the pancreatic tumour or a metastasis for six injections. The primary endpoints were safety and treatment-emergent immune response in patients who received at least one dose of LOAd703, and antitumour activity was a secondary endpoint. This study was registered with ClinicalTrials.gov, NCT02705196, arm 2 is ongoing and open to new participants.

Findings: Between Dec 2, 2016, and Oct 17, 2019, 23 patients were assessed for eligibility, leading to 22 patients being enrolled. One patient withdrew consent, resulting in 21 patients (13 [62%] men and eight [38%] women) assigned to a dose group (three to dose 1, four to dose 2, and 14 to dose 3). 21 patients were evaluable for safety. Median follow-up time was 6 months (IQR 4-10), and data cutoff was Jan 5, 2023. The most common treatment-emergent adverse events overall were anaemia (96 [8%] of 1237 events), lymphopenia (86 [7%] events), hyperglycaemia (70 [6%] events), leukopenia (63 [5%] events), hypertension (62 [5%] events), and hypoalbuminaemia (61 [5%] events). The most common adverse events attributed to LOAd703 were fever (14 [67%] of 21 patients), fatigue (eight [38%]), chills (seven [33%]), and elevated liver enzymes (alanine aminotransferase in five [24%], alkaline phosphatase in four [19%], and aspartate aminotransferase in four [19%]), all of which were grade 1-2, except for a transient grade 3 aminotransferase elevation occurring at dose 3. A maximum tolerated dose was not reached, thereby establishing dose 3 as the highest-evaluated safe dose when combined with nab-paclitaxel plus gemcitabine. Proportions of CD8⁺ effector memory cells and adenovirus-specific T cells increased after LOAd703 injections in 15 (94%) of 16 patients for whom T-cell assays could be performed. Eight (44%, 95% CI 25-66) of 18 patients evaluable for activity had an objective response.

Interpretation: Combining LOAd703 with nab-paclitaxel plus gemcitabine in patients with advanced pancreatic ductal adenocarcinoma was feasible and safe. To build upon this novel chemoimmunotherapeutic approach, arm 2 of LOKON001, which combines LOAd703, nab-paclitaxel plus gemcitabine, and atezolizumab, is ongoing.

Funding: Lokon Pharma, the Swedish Cancer Society, and the Swedish Research Council.

Figure 1.. Consort design.

^aReceived fewer than three doses of LOAd703. Abbreviations: VP, viral particles.

Figure 2.. Percentage change in the sum of target lesions and CA 19–9 levels compared to baseline in activity-evaluable subjects.

(A) Percentage change in the sum of lesions over time by LOAd703 dose level after treatment initiation (DL1 n=3 [lines for two subjects overlapped], DL2 n=4, DL3 n=11); (B) Maximum percentage change in sums of all target lesions (n=18), injected lesions (n=18), and non-injected lesions in individual subjects (n=15). Note that three subjects do not have any non-injected lesions (*). If more than one lesion was injected with LOAd703, the sum of all injected lesions was included in the calculation for best response; (C) Maximum percentage change in CA19–9 levels compared to baseline in each subject (DL1 n=3, DL2 n=4, DL3 n=11); and (D) Percentage changes in CA19–9 levels for each subject during the study over time since treatment initiation (n=18). Activity evaluable subjects received at least three doses of LOAd703. Abbreviations: PD, progressive disease; PR, partial response; SD, stable disease; VP, viral particles.

Figure 3.. Effects on immune cells during LOAd703 treatment.

(A) The proportion of adenovirus-specific T cells in peripheral blood was determined by ELISpot analysis pretreatment and after treatment initiation (w9/10) in safety-evaluable subjects (n=16, samples were lacking from 5 subjects). (B) Level of anti-adenovirus antibodies in serum were quantified over time by ELISA in safety-evaluable subjects (n=20, sample was lacking from 1 subject) and depicted per dose level. Each line represents an individual subject. Dotted lines represent subjectś re-treatment with LOAd703. Hash marks (#) indicate values above highest standard point. (C) The proportion of circulating CD8+ T cell subpopulations were quantified over time by flow cytometry in the activity-evaluable population (received at least 3 doses of LOAd703). Data are presented as fold-change relative to day 15 of cycle 1 (week 3) prior to treatment with LOAd703. Each line represents an individual subject (n=13, samples were lacking from 8 patients). Wilcoxon signed-rank test was used for calculations of statistical significant values compared to baseline. Abbreviations: VP, viral particle