CapRI: Final results of the open-label, multicenter, randomized phase III trial of adjuvant chemoradiation plus interferon-α2b (CRI) versus 5-FU alone for patients with resected pancreatic adenocarcinoma (PAC).

LBA4012 Background:Adjuvant chemomonotherapy in PAC results in five-year survival of 21% with median overall survival (mOS) of 23 months. Phase II trials evaluating adjuvant CRI showed promising results (mOS 27-44 months).,Patients with an R0/R1 resection for PAC were randomized <12 weeks of surgery to receive either 5-Fluorouracil (200mg/m/day, CI); Cisplatin (weekly 30mg/m) and 3 million units IFN-α (three times a week) for 5.5 weeks combined with external beam radiation (50.4Gy in 28 fractions) followed by two more cycles of continuous 5-FU or 5FU/FA (FA, 20 mg/m, iv bolus injection followed by 5-FU, 425 mg/m, iv bolus injection given 1-5d every 28 days) for 6 months. Patients treated with CRI were challenged prior to therapy with a single dose of IFN-α. The primary outcome measure was overall survival; the secondary measures were toxicity, progression free survival and quality of life. 110 patients were calculated to detect a difference in hazard on level α= 0.05 and with a power of 80%.,110 patients from five centers in Germany and Italy were randomized from July 2004 and December 2007. Median (range) age was 63 (33-77) years; 60 (57%) were men. 104 (95%) were T3 tumors, 87 (79%) were node positive and 43 (39%) were R1 resections, and 33 (30%) were poorly differentiated tumors. Grade 3 or 4 toxicity (mainly neutropenia) was observed in 68% of CRI and 16% of 5FU/FA (mainly diarrhea). Side effects during the multimodular cycle 1 were manageable and patients recovered completely. There was no difference in quality of life between the treatment groups. Final analysis was carried out on an intention to treat basis with a minimum of 2 years follow-up. Median survival of patients treated with 5FU/FA was 28.5 [95% CI: 19.5, 38.6] months and for patients treated with CRI this was 32.1 [95% CI: 22.8, 42.2] months. Although survival curves are clearly separated the log-rank analysis revealed no statistically significant difference in survival estimates. There was a clear trend towards better response for high risk patients (R1, start of treatment > 8 weeks after surgery; p=0.11). CRI reduced the risk of local recurrence (29.3% vs. 55.6%; p=0.014). Pre-planned testing for predictive markers showed that patients treated with CRI who responded to single IFN-α challenge with a decrease in T helper cells and especially regulatory T cells or with a pronounced increase in NK cell mediated cytotoxicity had a significantly longer survival.,This is the highest ever reported mOS for adjuvant PAC in a randomized trial. Unfortunately, this underpowered trial was not able to address the significance of CRI in PAC satisfactorily. There is evidence that especially high risk patients benefit from CRI; local control improved significantly. A strong immune response to a single IFN-α challenge is significantly associated with a good outcome. Confirmatory trials are needed. [Table: see text].