Three Drugs Marginally Better Than Two in First-Line Metastatic Colorectal Cancer

Johanna Bendell, MD

Patients with metastatic colorectal cancer (mCRC) had marginal yet inconsistent improvement in clinical outcomes when treated with a three-drug chemotherapy regimen plus bevacizumab compared with a doublet-bevacizumab regimen, according to results from the phase II STEAM trial.

About 60% of patients had objective responses (primary endpoint) with 5FU-leucovorin-oxaliplatin-irinotecan (FOLFOXIRI) and bevacizumab versus 47% with 5FU-oxaliplatin (FOLFOX) plus bevacizumab. The odds ratio for response of 1.7 in favor of FOLFOXIRI and concurrent bevacizumab (cFOLFOXIRI) did not achieve statistical significance, but the odds ratio of 1.8 for FOLFOXIRI and sequential bevacizumab (sFOLFOXIRI) did achieve significance, as did pooled data for the 2 treatment groups versus FOLFOX-bevacizumab.

For the second primary endpoint of progression-free survival (PFS), the cFOLFOXIRI regimen resulted in a statistically significant 2.4-month improvement in PFS versus FOLFOX, but sFOLFOXIRI’s 1.4-month improvement in PFS was not statistically better than the PFS associated with FOLFOX.

“Compared with FOLFOX and bevacizumab, the triplet regimen of concurrent FOLFOXIRI and bevacizumab demonstrated trends for improved objective response rate, progression-free survival, and metastatic resection rates,” said Johanna Bendell, MD, director of gastrointestinal cancer research at the Sarah Cannon Research Institute, when presenting the data at the 2016 Gastrointestinal Cancer Symposium.

“The present analysis of progression-free survival is preliminary, and updated results will be presented in the future. This still is now powered to compare overall survival results between treatment arms.”

The interim results are consistent with those of the previously reported TRIBE study (N Engl J Med. 2014;371:1609-1618), supporting FOLFOXIRI-bevacizumab as an option for first-line treatment of mCRC, she added.

Currently, FOLFOX in combination with bevacizumab is the most frequently used first-line systemic therapy for mCRC in the United States, Bendell noted. The TRIBE trial demonstrated significant improvement in objective response and PFS, and a trend toward improved overall survival with FOLFOXIRI-bevacizumab as compared with FOLFIRI (5FU-leucovorin-irinotecan) plus bevacizumab. TRIBE left unanswered the question of whether FOLFOXIRI-bevacizumab is superior to FOLFOX plus bevacizumab.

The randomized, open-label STEAM trial tried to answer this question. In the study, investigators randomized patients with previously untreated mCRC to one of 3 treatment groups: cFOLFOXIRI, sFOLFOXIRI, or FOLFOX-bevacizumab. Patients in all 3 groups received maintenance therapy with 5FU, leucovorin, and bevacizumab or capecitabine plus bevacizumab.

At first progression, all patients received second-line treatment with an investigator’s choice fluoropyrimidine-based regimen plus bevacizumab. Follow-up continued until second progression.

The primary endpoint was investigator-assessed objective response during first-line therapy for the cFOLFOXIRI-bevacizumab regimen versus FOLFOX-bevacizumab. The co-primary endpoint was PFS during first-line therapy with the FOLFOXIRI regimens versus FOLFOX.

The trial was statistically powered to detect an increase in objective response rate from 50% with FOLFOX to 70% with cFOLFOXIRI and an improvement in median first-line PFS from 10 months with FOLFOX to 14.3 months with the FOLFOXIRI regimens.

Data analysis comprised 280 randomized patients, who had a median follow-up of about 13 months. The patients had a median age of about 57, and 55% to 60% were men. Two-thirds of the patients had ECOG performance status 0, and 75% to 80% of the study population had colon cancer at diagnosis (as opposed to rectal). From 52% to 64% of patients across the 3 treatment groups had prior surgery.

About 30% of patients had liver-limited metastasis, left-sided tumor location was reported in 55% to 60% of cases, and KRAS wild-type in about 25%, KRAS mutant in about 25%, and KRAS unknown in the rest.

The overall response rates were 60.2% with cFOLFOXIRI, 62.0% with sFOLFOXIRI, and 61.1% for the two groups combined. The hazard ratio versus the FOLFOX arm achieved statistical significance for the sFOLFOXIRI (HR, 1.8; P = .040) and combined analysis (HR, 1.8; P = .025).

The median PFS in first line was 11.7 months with cFOLFOXIRI, 10.7 months with sFOLFOXIRI, and 9.3 months with FOLFOX. The cFOLFOXIRI regimen achieved statistical significance versus FOLFOX, represented by a hazard ratio of 0.672 and confidence intervals that did not cross 1.0. The sFOLFOXIRI regimen resulted in a PFS hazard ratio of 0.738 versus FOLFOX, associated with confidence intervals that overlapped, making the difference nonsignificant.

The pooled analysis of PFS also demonstrated a significant advantage for the FOLFOXIRI arms, resulting in a median PFS of 11.4 months (HR, 0.704; 95% CI, 0.537-0.923).

More patients in the FOLFOXIRI groups had successful resection of liver metastases (15.1% for cFOLFOXIRI, 9.8% for sFOLFOXIRI, and 12.4% combined) compared with the FOLFOX group (7.4%). However, none of the differences achieved statistical significance (P = .555 to P = .094).

Adverse event rates were similar across the 3 treatment groups, including all treatment-emergent events, grade 3+ adverse events, and withdrawals due to adverse events. The cFOLFOXIRI group was associated with a higher rate of study discontinuation secondary to treatment-emergent adverse events (15% vs 3% for sFOLFOXIRI and 6% for FOLFOX). The cFOLFOXIRI arm also had a higher incidence of grade 3/4 hypertension.

In conclusion, Bendel said, “The interim results of STEAM align with findings from TRIBE, supporting the use of a FOLFOXIRI-bevacizumab regimen in a select group of first-line mCRC patients.

Bendell JC, Tan BR, Reeves JA, et al. Overall response rate (ORR) in STEAM, a randomized, open-label, phase 2 trial of sequential and concurrent FOLFOXIRI-bevacizumab (BEV) vs FOLFOX-BEV for the first-line (1L) treatment (tx) of patients (pts) with metastatic colorectal cancer (mCRC). J Clin Oncol 34, 2016 (suppl 4S; abstr 492).

<<<

View more from the 2016 GI Cancer Symposium