American Society of Clinical Oncology 2010 colorectal update

Abstract

The 2010 American Society of Clinical Oncology (ASCO) Gastrointestinal (Colorectal) Cancer Track included several notable presentations. The addition of cetuximab to FOLFOX in stage III colon cancer did not improve disease-free survival, but increased toxicity. In the metastatic setting, cetuximab demonstrated benefit only in a small subset of patients (KRAS wild-type and limited metastatic disease). Bevacizumab monotherapy may be equivalent to combination chemotherapy in the maintenance phase of treatment in advanced disease, and in another study bevacizumab did not appear to incur excess morbidity in patients with an intact primary tumor. Alternate strategies for the treatment of stage II/III rectal cancer included short-course radiotherapy with adjuvant chemotherapy and neoadjuvant FOLFOX–bevacizumab without radiation, both demonstrating promising results.

Colorectal cancer (CRC) ranks third in incidence as well as cancer-related mortality for both men and women in the USA, with an estimated 146,970 new cases and 49,920 deaths in 2009. Overall, 5-year survival from this disease is 65%, but for the 19% of patients who are diagnosed with metastatic disease, 5-year overall survival is less than 10% [1]. The colorectal proceedings at the 2010 American Society of Clinical Oncology (ASCO) meeting continued to provide insight into cytotoxic and biologic therapy and mechanisms of resistance to improve patient survival from this disease.

Adjuvant cetuximab for stage III colon cancer

The addition of cetuximab to FOLFOX in the adjuvant setting does not improve survival for patients with stage III colon cancer, according to results presented from the North Central Cancer Treatment Group (NCCTG) Intergroup Phase III N-0147 trial [2]. The trial randomized patients with stage III disease to 12 cycles of mFOLFOX-6 versus 12 cycles of mFOLFOX-6 plus cetuximab, with the primary end point of disease-free survival (DFS). In 2008, the trial underwent interim redesign to incorporate prospective KRAS testing and enrolled patients without mutations in KRAS from that point forward. The two arms demonstrated overlapping survival curves with a DFS hazard ratio of 1.2 (p = 0.22). On subset analysis, the addition of cetuximab was ineffective across all subgroups studied. The cetuximab arm had a 71% grade 3–4 toxicity rate compared with 51% in the control arm, and was more toxic in patients over 70 years of age.

Follow-up analysis of patients who harbored the KRAS mutation but who were randomized prior to study redesign in 2008 was presented in a separate analysis [3]. Slightly worse DFS was demonstrated in the cetuximab arm (64.2 vs 67.2% 3-year DFS; HR: 1.2; p = 0.13), but notably there was lower DFS in both KRAS mutant arms when compared with KRAS wild-type – a result suggesting a negative prognostic significance for the KRAS mutation in stage III disease.

Cetuximab in advanced CRC

Data presented from the mature Medical Research Council COIN trial showed a surprising lack of benefit to cetuximab in the metastatic setting, and stand in contrast to the Cetuximab Combined with Irinotecan in First Line Therapy for Metastatic Colorectal Cancer (CRYSTAL) and Oxaliplatin and Cetuximab in First-Line Treatment of Metastatic Colorectal Cancer (OPUS) studies [4]. In this trial, patients were given infusional 5-fluorouracil (5-FU) or capecitabine plus oxaliplatin, with or without cetuximab, and continued until progression, toxicity or patient discretion. All patients went on to receive irinotecan-containing therapy in the second line. The primary end point was overall survival in KRAS wild-type patients, with secondary end points of survival in KRAS mutant, ‘all wild-type’ (KRAS, NRAS and BRAF wild-type) and any mutant populations. Surprisingly, overall survival in the KRAS wild-type cetuximab-containing arm was 17 versus 17.9 months in the control arm; in both arms, progression-free survival was 8.6 months.

From the mature data set, the investigators sought to identify potentially responsive subsets of patients. KRAS wild-type patients who harbored zero or one metastatic site(s) (the former group comprised of patients with local recurrence) and who received infusional 5-FU therapy derived small benefit from the addition of cetuximab, a finding that requires prospective, randomized validation. This discrepancy may be due in part to higher rates of grade 3–4 diarrhea in patients receiving capecitabine, resulting in a global dose reduction of therapy. Whether oxaliplatin is indeed the ideal chemotherapy platform in which to pursue the addition of biologic therapies is a question requiring further study.

Of note, a negative prognostic effect of any activating mutation, independent of cetuximab use, was observed. While BRAF has been demonstrated in multiple settings to incur a poor prognosis, this observation has been less clear for KRAS – a finding perhaps attributable to prior analyses of KRAS confined to the third-line setting when compared with best supportive care. Upon initiation of first-line chemotherapy, it is emerging that KRAS may indeed confer a negative prognostic, as well as predictive, value.

Bevacizumab in the advanced CRC setting

The Spanish Cooperative Group MACRO trial evaluated the role of bevacizumab (Bev) maintenance after induction chemotherapy for patients with advanced CRC [5]. Patients with performance status 0–2 and no prior therapy for advanced disease received six cycles of capecitabine, oxaliplatin and Bev (XELOX/Bev) given every 3 weeks, followed by randomization of maintenance therapy to continued XELOX/Bev versus Bev monotherapy. A noninferiority study, the primary end point was defined as progression-free survival, but allowed for wide confidence limits. Progression-free survival in the Bev-only arm was 9.7 versus 10.4 months in the combination therapy arm, conferring a hazard ratio of 1.11. However, the confidence interval for this hazard ratio extended to 1.37, which at the extreme limit translates into a PFS benefit of 3.7 months for continued chemotherapy – an observation that warrants further confirmation of the noninferiority of Bev alone after a relatively short period of induction chemotherapy. Other criticisms include the lack of an observation control arm and assessment of time to chemotherapy resistance. Several prospective analyses of maintenance Bev are ongoing in the Double Reintroduction with Erlotinib and Avastin in Metastatic Colorectal Cancer (DREAM) and CAIRO3 trials, and should further clarify this important issue.

Recently, there has been a shift away from resecting the asymptomatic primary tumor in patients with synchronous unresectable disease, and an analysis of the NSABP C-10 trial contributed important safety information regarding Bev use in this setting [6]. With a primary end point of major morbidity (defined as obstruction, perforation, bleeding or death), this single-arm Phase II trial enrolled patients with an asymptomatic, unresected primary to FOLFOX6 plus Bev in a Simon two-stage design, with stage I powered to rule out a 40% rate of major morbidity. In this cohort, there was a 14% major morbidity rate, suggesting preliminarily that intact primary tumors and Bev use are not incompatible.

Optimal neoadjuvant therapy for rectal cancer

Finally, several ASCO 2010 abstracts focused on the optimal neoadjuvant management of T3 and T2 node-positive rectal cancer. In the Australia–New Zealand intergroup trial presented by Ngan et al., patients with clinical T3 tumors were randomized to short-course radiotherapy (25 Gy in five fractions) versus long-course preoperative chemoradiotherapy (50.4 Gy + CI 5-FU) [7]. Importantly, all patients went on to receive adjuvant chemotherapy. Powered to detect differences in local control rates as a primary end point, the study found no significant difference in local recurrence in a relatively short follow-up period (median follow-up: 3.5 years). There were no differences in the rate of distant metastases or overall survival. Longer follow-up is necessary to determine that late toxicity and recurrences are not different.

An alternative approach to neoadjuvant management with chemotherapy alone was explored by Schrag et al.[8]. In this pilot feasibility study, 29 patients with uT2N+ or uT3N0–1 disease who were candidates for sphincter-sparing surgery received six cycles of FOLFOX with Bev in cycles 1–4, with the primary end point of R0 resection rate. While patients were planned for further preoperative therapy based on progression/disease stability by imaging criteria, notably 29 out of 29 patients had clinical regression and proceeded to surgery without preoperative radiotherapy, and all 29 had R0 resections. The pilot study demonstrated a 27% pathologic complete response rate, which compares favorably with historical data of preoperative chemoradiation. Local and distant recurrence rates and DFS (secondary study end points) are immature, but based on this preliminary data, further study of nonradiotherapy approaches to neoadjuvant therapy are warranted.

While the data presented at this year’s colorectal session did not have immediate practice-changing implications, overall the colorectal session continued to further our understanding of the uses and limitations of biologic therapy to manage advanced disease. Ongoing trials will shed light on some of the most important management issues, including the role of maintenance Bev and the use of first-line cetuximab in the advanced CRC setting.

Financial & competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.