TGF-β is an immunosuppressive cytokine that is upregulated in colorectal cancer. TGF-β blockade improved response to chemoradiotherapy in preclinical models of colorectal adenocarcinoma. We aimed to test the hypothesis that adding the TGF-β type I receptor kinase inhibitor galunisertib to neoadjuvant chemoradiotherapy would improve pathological complete response rates in patients with locally advanced rectal cancer.

This was an investigator-initiated, single-arm, phase 2 study done in two medical centres in Portland (OR, USA). Eligible patients had previously untreated, locally advanced, rectal adenocarcinoma, stage IIA–IIIC or IV as per the American Joint Committee on Cancer; Eastern Cooperative Oncology Group status 0–2; and were aged 18 years or older. Participants completed two 14-day courses of oral galunisertib 150 mg twice daily, before and during fluorouracil-based chemoradiotherapy (intravenous fluorouracil 225 mg/m² over 24 h daily 7 days per week during radiotherapy or oral capecitabine 825 mg/m² twice per day 5 days per week during radiotherapy; radiotherapy consisted of 50·4–54·0 Gy in 28–30 fractions). 5–9 weeks later, patients underwent response assessment. Patients with a complete response could opt for non-operative management and proceed to modified FOLFOX6 (intravenous leucovorin 400 mg/m² on day 1, intravenous fluorouracil 400 mg/m² on day 1 then 2400 mg/m² over 46 h, and intravenous oxaliplatin 85 mg/m² on day 1 delivered every 2 weeks for eight cycles) or CAPEOX (intravenous oxaliplatin 130 mg/m² on day 1 and oral capecitabine 1000 mg/m² twice daily for 14 days every 3 weeks for four cycles). Patients with less than complete response underwent surgical resection. The primary endpoint was complete response rate, which was a composite of pathological complete response in patients who proceeded to surgery, or clinical complete response maintained at 1 year after last therapy in patients with non-operative management. Safety was a coprimary endpoint. Both endpoints were assessed in the intention-to-treat population. This study is registered with ClinicalTrials.gov, NCT02688712, and is active but not recruiting.

Between Oct 19, 2016, and Aug 31, 2020, 38 participants were enrolled. 25 (71%) of the 35 patients who completed chemoradiotherapy proceeded to total mesorectal excision surgery, five (20%) of whom had pathological complete responses. Ten (29%) patients had non-operative management, three (30%) of whom ultimately chose to have total mesorectal excision. Two (67%) of those three patients had pathological complete responses. Of the remaining seven patients in the non-operative management group, five (71%) had clinical complete responses at 1 year after their last modified FOLFOX6 infusion. In total, 12 (32% [one-sided 95% CI ≥19%]) of 38 patients had a complete response. Common grade 3 adverse events during treatment included diarrhoea in six (16%) of 38 patients, and haematological toxicity in seven (18%) patients. Two (5%) patients had grade 4 adverse events, one related to chemoradiotherapy-induced diarrhoea and dehydration, and the other an intraoperative ischaemic event. No treatment-related deaths occurred.

The addition of galunisertib to neoadjuvant chemoradiotherapy in patients with locally advanced rectal cancer improved the complete response rate to 32%, was well tolerated, and warrants further assessment in randomised trials.

Eli Lilly via ExIST program, The Providence Foundation.