Clinical manufacture of CRISPR/Cas9-based cytokine-induced SH2 protein knock-out tumor-infiltrating lymphocytes for gastrointestinal cancers

Abstract

Introduction

The prognosis of stage IV gastrointestinal (GI) carcinomas is poor with a 15% five-year survival rate for colorectal carcinomas. To improve efficacy of tumor infiltrating lymphocytes (TIL), we isolated mutation-reactive autologous TIL and employed CRISPR/Cas9 to knockout (KO) the intracellular checkpoint protein CISH, which has been shown to enhance T cell expansion, functional avidity, and cytokine polyfunctionality, with consequent durable regression of established tumors in an animal model.

Materials & Methods

TIL cultures were initiated from resected tumor fragments and maintained for six weeks before harvest and cryopreservation. Candidate neoantigens were nominated by exome sequencing and peptides were used to identify mutation reactive (MR) TIL. Selected MR TIL were thawed and allowed to recover for 24–36 h in media with 10% AB serum, 6000 IU/mL IL-2, and 5 ng/mL IL-7 and IL-15 followed by stimulation with plate-bound anti-CD3/soluble anti-CD28 for 4 days. CISH KO was performed by electroporation of Cas9 mRNA and chemically modified single guide RNA. Between 5 -7.5 million viable cells were added to each 100 cm² G-Rex vessel containing 600 mL expansion media (with allogeneic feeder MNC:TIL = 100:1) and incubated for 6–8 days. Cultures were evaluated and split according to cell concentration criteria (and dose cohort) and incubated for an additional 6-8 days. On day 14, all of the cells were harvested, washed with buffer and cryopreserved (5% DMSO). Lot release testing included: viability, %CD3+, cytology review, Gram stain, sterility, endotoxin, mycoplasma, and interferon gamma (IFN-γ) production. Additional testing included DNA sequencing to determine genomic CISH editing efficiency and a Western blot for determination of CISH protein loss.

Results

Patients with GI cancers (colon [10], rectal [8], pancreatic [1], and esophageal [1]) underwent tumor collection. Nineteen of 22 tumor biopsies sampled from 20 patients total proceeded to KO/expansion. Final TIL product results (mean [SD], median [range]) were: viable count (x 10¹⁰) -3.25 (3.67), 1.95 (0.018–12.40); viable TIL fold expansion -327.1 (364.8), 153.1 (8–1454); % viability - 76 (13), 78 (43–92); % CD3 -94.4 (5.4), 95.8 (78.6–99.4); % CISH KO efficiency – 75 (29), 87 (0–96); % editing efficiency - 59.9 (24.8), 66.9 (0.4–86). Viability fell below 70% for five TIL products. All other lot release testing has met specification. Thirteen patients have received TIL; six patients were not treated due to disease progression prior to anticipated infusion.

Conclusion

The translation of CRISPR/Cas9-based CISH KO MR TIL from the basic research lab to current good manufacturing practices The (cGMP) facility was successful, allowing for optimized, large-scale expansion in support of a first-in-human clinical trial to treat patients with metastatic GI cancers (ClinicalTrials.gov Identifier: NCT04426669).