Is the paradigm wrong?

Adjuvant chemotherapy improves survival in patients with stage III colon cancer, but as long as the risk of recurrence remains, our job is not done.

The biologic agents bevacizumab (Avastin, Genentech) and cetuximab (Erbitux, ImClone) were obvious candidates to add to cytotoxic chemotherapy in these patients: Why wouldn’t the marginal benefit seen in advanced disease with these combinations translate into cures in the adjuvant setting? Thus, the profound disappointment when the trials investigating bevacizumab or cetuximab with FOLFOX in the adjuvant setting both failed to demonstrate a survival advantage. First, the post-mortem and then the question: Where to now?

Bevacizumab

Angiogenesis is a key factor in the continued growth of tumors and is an attractive target for cancer therapeutics. Bevacizumab is a monoclonal antibody directed against VEGF, the dominant growth factor involved in angiogenesis. It was, of course, the first antiangiogenesis agent approved in the treatment of cancer after demonstrating improved survival when used in conjunction with bolus 5-FU/leucovorin and irinotecan in patients with metastatic colon cancer. Subsequent studies pairing bevacizumab with the more standard irinotecan-based regimens, as well as oxaliplatin-based chemotherapy, also confirmed favorable outcomes. Based on this success in the metastatic setting, several trials were designed to test the role of bevacizumab in the adjuvant setting.

The national Surgical Adjuvant Breast and Bowel Project conducted NSABP C-08, a phase-3 trial with 2,710 patients with stage II or III colon cancer. Patients were randomly assigned, after curative resection, to receive either modified FOLFOX-6 alone for six months or modified FOLFOX-6 for six months plus bevacizumab for a total of 12 months. The results of this study were presented at the plenary session of the 2009 ASCO Annual Meeting. Simply stated, there was no significant improvement in three-year DFS — the primary endpoint — in the patients who were assigned bevacizumab. There was an apparent improvement in DFS seen during the year in which patients received bevacizumab that vanished over time. This suggests the possibility of a transient beneficial effect of the additional six months of bevacizumab, although this observation may be overstated due to an asymmetry in the scanning intervals between the two arms.

Anne M. Espinoza

The AVANT trial is a randomized phase-3 trial investigating the addition of bevacizumab to FOLFOX-4 and XELOX in 3,450 patients with stage II and III colon cancer. We await the occurrence of enough events to trigger an efficacy analysis from this trial to see if the findings of C-08 are validated or refuted. However, the clear-cut results of C-08 make it unlikely that AVANT will show much, if any, advantage to bevacizumab in terms of three-year DFS (although the mandatory scan intervals in AVANT will inform the question of a possible transient benefit of bevacizumab).

Cetuximab

Cetuximab is a monoclonal antibody that binds to epidermal growth factor receptor. The EGFR signaling pathway is involved in several important steps in tumor progression, including cell proliferation, angiogenesis and spread of metastases. Cetuximab has demonstrated activity in the treatment of metastatic colon cancer, both when used as monotherapy in patients who have failed standard therapies and when added to standard oxaliplatin or irinotecan-based regimens. Based on these data, a phase-3 trial assessing the use of cetuximab in the adjuvant setting was undertaken.

The study was amended after it became clear that only patients whose tumors did not harbor KRAS mutations benefited from the EGFR antibody.

Thus, after much iteration, the North Central Cancer Treatment Group (NCCTG) trial NO147 ultimately tested FOLFOX alone vs. FOLFOX with cetuximab in stage III patients with KRAS wild-type colon cancers. Unfortunately, in a recently performed preplanned interim analysis, it was determined that the addition of cetuximab demonstrated no benefit, nor was it statistically plausible to show benefit with further time and more patients. Based on these findings, the study has been permanently closed.

Why and where do we go from here?

The paradigm for therapy development has been to apply therapies successful in metastatic patients to the adjuvant treatment of stage II and III patients; thus, the logical trials described above. But alas, these two large trials are clearly negative and raise a series of questions. Although the results come as a surprise, we must ask: Does the science behind this approach make sense? And, if so, why aren’t we seeing the results we expected? And where do we go from here?

Alan P. Venook

For bevacizumab, is it possible that micrometastases that remain after surgical resection grow independent of tumor-specific vasculature, allowing growth despite the presence of an antiangiogenesis agent? Or, based on the possibility of a transient benefit seen with the additional six months of bevacizumab in NSABP C-08, is longer exposure to bevacizumab required to demonstrate improved efficacy? This idea has led NSABP to propose C-12 as a successor to C-08, which would incorporate an additional year of treatment with bevacizumab. However, the question remains, how long would be long enough, and do the added toxicities, unknown possible long-term complications and cost justify the possible benefit?

The failure of the addition of cetuximab to improve outcomes in N0147 is perhaps more distressing, given the ability to employ a marker that predicts response — KRAS mutational status — to enrich the patient population. Also, as opposed to bevacizumab, cetuximab has monotherapy efficacy that might have mechanistically favored its additive role in the adjuvant setting. Nonetheless, the study failed, and there is no hint of efficacy that will lead to another new study with the EGFR antibodies in the adjuvant setting.

And the next step? The Intergroup will ask an oxaliplatin duration question — six vs. 12 cycles of FOLFOX — but will add a “biologic” of a different sort — celecoxib (Celebrex, Pfizer) — in a second randomization. Given the flash of the two failed studies, it would be fitting if this less costly and relatively stodgy question gets the right answer.

Anne M. Espinoza, MD, is a Fellow in Hematology/Oncology at the University of California, San Francisco.

Alan P. Venook, MD, is Professor of Clinical Medicine and Associate Chief, Division of Medical Oncology, University of California, San Francisco and is also a member of the HemOnc Today Editorial Board.

For more information:

• Bokemeyer C. J Clin Oncol. 2009;27:663-671.
• Cunningham D. N Engl J Med. 2004351:337-345.
• Fuchs CS. J Clin Oncol. 2008;26:689-690.
• Giantonio BJ. J Clin Oncol. 2007;25:1539-1544.
• Hochster HS. J Clin Oncol. 2008;26:3523-3529.
• Hurwitz H. N Engl J Med. 2004;350:2335-2342.
• Jonker DJ. N Engl J Med. 2007;357:2040-2048.
• Saltz LB. J Clin Oncol. 2008;26:2013-2019.
• Sobrero AF. J Clin Oncol. 2008;26:2311-2319.
• Van Cutsem E. Ann Oncol. 2009;20:1842-1847.
• Van Cutsem E. N Engl J Med. 2009;360:1408-1417.
• Wolmark N. #LBA4. Presented at: ASCO 2009 Annual Meeting; May 29-June 2, 2009; Orlando.