The oncolytic adenovirus TILT-123 with pembrolizumab in platinum resistant or refractory ovarian cancer: the phase 1a PROTA trial

Abstract

Immune checkpoint inhibitors have demonstrated modest efficacy as a monotherapy in ovarian cancer. Originally developed to improve efficacy of T-cell therapies such as immune checkpoint inhibitors and adoptive cell transfer, TILT-123 (Ad5/3-E2F-D24-hTNFα-IRES-hIL-2) is a serotype chimeric oncolytic adenovirus encoding tumor necrosis factor alpha and interleukin-2. Here we report results from phase 1a of PROTA, a single-arm, multicentre dose escalation trial with TILT-123 and pembrolizumab in female patients with platinum resistant or refractory ovarian cancer (NCT05271318). The primary endpoint was safety. Secondary endpoints included efficacy, tolerability, virus persistence and anti-viral immunity. Patients (n = 15) received intravenous and intraperitoneal and/or intratumoral injections of TILT-123 as well as intravenous pembrolizumab. Treatment was well tolerated, and no dose-limiting toxicities were observed. The most frequent adverse events were fever (40%), fatigue (40%) and nausea (40%). Disease control was achieved in 64% of evaluable patients (9/14). Median progression-free survival and overall survival were 98 and 190 days respectively. Clinical responses were associated with higher serum anti-adenovirus neutralizing antibody titer at baseline and post-treatment. The phase 1b investigating TILT-123, pembrolizumab and PEGylated liposomal doxorubicin in a similar patient population is underway.

Fig. 1. Summary of treatment responses.

a Trial outline. b Median Overall survival. c Median Progression-free survival. d Treatment schedule (Created in BioRender. Clubb, J. [2024] https://BioRender.com/x81d347). e Spider plot showing sum of target-lesion response, displayed by best response (RECIST1.1) and evaluated by CT on day 36 and 92. f Waterfall plot showing sum of target-lesion response, displayed by response (RECIST1.1) at end of trial. Data cut-off 16/05/2024. Source data are provided as a Source Data file.

Fig. 2. Examples of treatment induced changes in tumour size.

a CT scan of a injected lesion (circled in red) in the left lung of patient 301-05, before (left) and after (right) treatment on day 36, indicating a 35.8% reduction in tumor diameter from 12 mm to 7.7 mm following three injections of TILT-123 only. b CT scan of injected and non-injected lesions in medial abdominal wall (top row) and liver metastasis (bottom row) of patient 301-11. CT scans display before treatment (left) and after treatment on day 36 (TILT-123 only) (middle) and day 92 (TILT-123 and pembrolizumab) (right). CT value change from baseline to day 92 of injected abdominal wall metastases indicates a 40% reduction in tumor diameter from 30 mm to 18 mm, whilst the non-injected liver metastasis shows a 33.3% reduction in tumor diameter from 30 mm to 20 mm.

Fig. 3. Pharmacokinetics, virus detection and immunomodulation of tumors.

a Concentration of TILT-123 present in whole blood of patients in cohorts 2–4 during the trial, as determined by qPCR (n = 7). b Detection of TILT-123 in injected and non-injected tumors as determined by quantification of adenovirus hexon and fiber knob mRNA and TILT-123 transgenes IL2 and TNF (n = 6). c Detection of TILT-123 in non-injected tumors of patient 302-06 at baseline compared to day 78, as determined by anti-hexon immunohistochemistry, where areas of interest are boxed in yellow and red and black arrows indicate hexon positive (brown DAB stain) adenovirus (TILT-123) positivity. d Characterisation of an injected tumor from patient 302-06 showing basophilic inclusions in white dotted circles (top row) which co-localise with hexon (brown DAB stain) positive stain indicated by black arrows (lower row). Right quadrant shows microcalcifcations within tumor nests circled in black dotted circles which co-localise with hexon positivity indicated by black arrows. Magnified area of interest is shown on the right column of the quadrant. e Multiplex immunofluorescence of the matching biopsies above in figure c, indicating changes in CD8+, CD4+, CD20+ and PD-L1+ expressing cells from baseline compared to day 78. Matching areas of interest boxed in yellow and red. Data sets in a and b are presented as mean ± SEM and significance represented by exact p-values. Comparisons was evaluated two-tailed Mann Whitney U test. n number of patients (biological replicates). Source data are provided as a Source Data file.

Fig. 4. Immunomodulation of tumors and neutralizing antibodies as a potential biomarker.

a Percentage (upper row) and fold change (lower row) during treatment in injected and non-injected tumors for intraepithelial PD-1 expressing CD8 T cells, intraepithelial PD-1 expressing CD56 NK cells, intraepithelial CD4 T cells and all CD20 B cells present in biopsy (n = 6). b Neutralizing antibody titers generation across trial, stratified by cohort and best clinical response (RECIST1.1) with red indicating progressive disease (PD) and dark green indicating stable disease (SD) and light green indicating partial response (PR). c Association of best clinical response (RECIST1.1) and anti-adenovirus nAb titer developed following treatment with TILT-123 (n = 14). d Association of best clinical response and presence of nAbs at baseline line (e). Association of overall survival with anti-adenovirus nAb titer developed following treatment (n = 14). f Association of overall survival with presence of nAbs at baseline. Percentage is presented as mean ± SEM and significance represented as *p < 0.05, **p < 0.001. Comparison between groups was evaluated by two-tailed Mann Whitney U test or Welch’s t test. Overall survival in e was evaluated using Log-rank (Mantel-Cox) test and Mann Whitney U test for overall survival in f. n number of patients (biological replicates). Source data are provided as a Source Data file.