Introduction

The past decade has been an era of rapid proliferation of new drugs and regimens for metastatic colorectal cancer (mCRC).  With seven active drugs now available including four chemotherapeutic agents: 5-fluorouracil, capecitabine, irinotecan, and oxaliplatin, and three targeted agents: bevacizumab, cetuximab, and panitumumab, there are many possibilities for combinations.  The challenge now is to identify how best to use all effective agents to maximum advantage through the course of treatment for this disease.  Presentations at ASCO 2007 began to deliver answers to these questions. 

Addition of Biologicals to Chemotherapy

There is no doubt that the introduction of the antiangiogenic agent bevacizumab has transformed the treatment of mCRC.  The benefit of bevacizumab was first identified in combination with the IFL regimen.  Results with bevacizumab added to more current chemotherapy programs have being awaited.  An important first-line trial, the BICC-C study, was updated at ASCO 2007.[1] This trial, initially designed as a three-arm study of FOLFIRI versus modified bolus IFL versus capecitabine plus irinotecan (Cape/IRI) +/- celecoxib, was amended in April 2004 to add bevacizumab to the FOLFIRI and modified bolus IFL arms and to discontinue the Cape/IRI arm due to increased toxicity from this regimen without increased activity. Therefore, there are patients in this trial treated both pre and post the addition of bevacizumab. This report updated results from both groups of patients.  

All patients had good performance status (0-1) and only 9-10% had prior chemotherapy.  The primary endpoint was progression free survival (PFS).  Celecoxib did not improve either PFS or overall survival.  There was a trend in the pre-bevacizumab patients for FOLFIRI being the superior regimen in terms of both PFS and OS.  Survival was improved as well with 87% one-year survival on FOLFIRI plus bevacizumab versus 61% in modified bolus IFL plus bevacizumab.  The median OS for FOLFIRI plus bevacizumab has not yet been reached. Median PFS for FOLFIRI plus bevacizumab was 11.2 months versus 8.3 months for IFL plus bevacizumab (HR 1.4, p=0.28).  

Bevacizumab added little toxicity to FOLFIRI except grade 3/4 neutropenia (53.6% vs. 43%), hypertension (12.5% vs. 2.2%),  hand-foot syndrome (3.6% vs. 0%), and pulmonary embolism (3.6% vs. 2.2%).  The sixty day mortality was 3.6% for FOLFIRI and 1.8% for FOLFIRI plus bevacizumab.  It should be kept in mind these cohorts were sequential rather than concurrent. 

What about the regimen most commonly used in the United States for the treatment of metastatic colorectal cancer - FOLFOX plus bevacizumab?  This regimen was adopted without the benefit of first-line data after impressive results were provided in the second-line setting. Dr. Saltz presented an update of the XELOX-1 trial, an open label study initially comparing capecitabine plus oxaliplatin (XELOX) to FOLFOX 4.[2]   In 2004, this study too was amended to test the addition of bevacizumab with an adjustment in trial design to a partially blinded 2 x 2 factorial design.  

Previously reported data from this trial demonstrated that XELOX was non-inferior to FOLFOX and these results were updated at ASCO 2007, demonstrating virtually identical OS of 19.6 and 19.8 months.[3]   The combined analysis of XELOX and FOLFOX with or without  bevacizumab showed improvement in PFS for the addition of the antiangiogenic agent, 9.4 months versus 8.1 months (HR 0.83, p<0.0001).  The median time to progression was 6.9 months versus. 6.0 months (p=0.003, HR 0.84) and there was a trend towards improved overall survival, 21.3 months versus 19.9 months, p=.0769.  Interestingly, the response rate was nearly identical in both arms.  Median treatment duration was similar across all arms.  

The incidence of grade 3/4 toxicity was comparable between bevacizumab and placebo arms.  However, more patients receiving bevacizumab discontinued treatment due to adverse events while more discontinued therapy in the chemotherapy alone arms from disease progression. The authors also did a comparison of patients who were in the intent-to-treat group, many of whom had discontinued treatment versus a group of patients who stayed on treatment until progression and found a difference in the PFS for the bevacizumab (10.4 months vs. 7.9 months, p=<0.0001) suggesting continuing bevacizumab until progression is the best strategy.  

Other studies addressed the addition of bevacizumab to FOLFIRI.  The AVIRI trial, a Phase 4 study enrolled 209 patients to this regimen.[4] The response rate was 53% and the PFS was 11.1 months median.  A smaller, single institution Phase 2 study reported on 42 patients treated with FOLFIRI plus bevacizumab.[5] The response rate was 61% with a median PFS of 12.0 months.  The median overall survival has not yet been reached although follow up for this trial remains short.  Grade 3/4 toxicity included neutropenia 40%, hypertension 19% and fatigue.  Pulmonary embolism developed in 4 patients.  Schmiegel reported a randomized Phase II study comparing CapeOX + bevacizumab to CapeIRI + bevacizumab in mCRC.[6] Response rates were similar, 45 versus 48% OS, while preliminary PFS rates at 6 months are 74 versus 80%.  Grade 3/4 toxicities demonstrated more sensory neuropathy for CapeOX but were otherwise similar. 

Taken together these trials demonstrate that FOLFIRI, FOLFOX or XELOX plus bevacizumab are effective first-line regimens with apparently similar PFS and response rates. Baseline patient characteristics and perhaps prior chemotherapy may substantially influence outcome to this general therapeutic approach. While toxicity is increased somewhat by bevacizumab, it does appear to be easily manageable and most patients with metastatic colorectal cancer should benefit from bevacizumab containing first-line regimens. 

Another trial of great interest was the Phase III CRYSTAL study which compared FOLFIRI +/- cetuximab in the first-line treatment of metastatic colorectal cancer.[7] Dr. Van Cutsem presented the results based on 1,217 patients enrolled and 1,198 patients eligible for an intent-to-treat analysis.  Baseline and disease characteristics were well-balanced between arms.  In general, this was a favorable patient population with less than 20% receiving prior adjuvant chemotherapy, the majority with ECOG 0-1 and 80% had < two metastatic sites. The primary endpoint, PFS was statistically significantly improved by the addition of cetuximab to 8.9 months compared to 8.0 months, HR 0.85, p=0.0479.  Although the clinical improvement in median PFS of under one month is modest, the one-year PFS rate was 34% for FOLFIRI plus cetuximab vs. 23% for FOLFIRI alone, and the independently-assessed response rate increased to 47% versus 39%, p=0.005.  Overall survival data are not yet mature and were not reported.  

The authors also reported on surgery with curative intent after FOLFIRI + cetuximab.   Not only did a statistically greater number of patients receiving cetuximab undergo surgical resection, 6 vs. 2.5%, more patients who received cetuximab had no evidence of disease at surgery, 4.3% versus 1.5%, p=0.0034.  An exploratory analysis suggested the addition of cetuximab doubled the complete resection rate in patients with liver-confined disease from 4.5% to 9.8%. 

Toxicity in the CRYSTAL trial was predictable with common grade 3/4 toxicities being skin rash (18.7% vs. 0.2%), diarrhea (10.2% vs. 10.5%) and neutropenia (20.7% vs. 20.3%).  The overall incidence of grade 3/4 events was 78% vs. 60% in the FOLFIRI alone arm.  Thus, this trial establishes FOLFIRI plus cetuximab as another option in the first-line setting, perhaps particularly beneficial for patients intending to proceed with liver resection after initial chemotherapy. 

Dr. Bokemeyer presented results of the OPUS trial, a Phase II study comparing FOLFOX to FOLFOX plus cetuximab.[8] The study results are still preliminary but the addition of cetuximab increased the response rate by 10% from 35.7% to 45.6%.  Virtually all subgroups benefited from the addition of cetuximab.  An association between grade of skin rash and response rate was noted.  These studies suggest that cetuximab can be combined with standard first-line chemotherapy approaches to improve clinical benefit. Whether bevacizumab or cetuximab, or both together, coupled with chemotherapy in the first-line setting is the best approach in mCRC will await forthcoming clinical trials addressing these questions. 

Chemotherapy Regimens

There remains uncertainty regarding the best way to sequence the active chemotherapeutic agents. The Dutch Colorectal Cancer Group designed a trial called the CAIRO study to determine whether combination therapy is more effective as the first-line chemotherapeutic approach compared to sequential treatment with the same agents.[9] Trial designs of this nature are often problematic as they require patients to have second-line and subsequent salvage regimens defined prospectively.  In this study, patients received either sequential therapy: first-line capecitabine followed by second-line irinotecan followed by third-line capecitabine plus oxaliplatin, or combination therapy: capecitabine plus irinotecan first-line followed by capecitabine plus oxaliplatin second-line.  Drug dosages were kept as close as possible to equality between the arms.  

Of 820 patients randomized, 803 were eligible and 795 received at least one cycle of treatment.  The primary endpoint of the trial was overall survival. Baseline characteristics were well balanced, with almost all patients having a performance status of 0-1.    There was no significant difference between arms for OS (HR 0.92, p=NS).  The median survival was 16.3 months in the sequential arm vs. 17.4 months in the combination arm (p=0.33).  As anticipated, the response rate was doubled in the combination arm, 41% vs. 20%, although second-line response rates were similar despite comparing a singlet in the sequential arm to a doublet in the combination arm. Median PFS was 5.8 months for sequential versus 7.8 months for combination (p=0.0002).  

Quality of life scores were similar across arms while toxicity was generally greater in the combination arm with grade 3/4 toxicity rates for nausea, vomiting and febrile neutropenia significantly higher in the combination arm compared to the sequential (although <10% absolute in the sequential arm), 26% diarrhea versus 11% in sequential.  Grade 3/4 hand-foot syndrome was higher in the sequential, 12 % versus 6%.  

The authors concluded that sequential therapy with these agents is a viable option with similar survival to combination therapy.  It remains to be determined whether targeted therapies would alter these results; and consequently the treatment strategy.  Until such studies are conducted, these results while interesting, have limited impact on current practice patterns. 

Chemotherapy-Free Interval

Another issue of interest recently is whether or not metastatic colorectal cancer patients benefit from a chemotherapy-free interval (CFI).  The rationale for this concept stems from the fact that mCRC can now be considered a chronic disease and the intent of therapy is to maximize length of life while simultaneously preserving or enhancing quality of life.  The OPTIMOX-1 trial, previously reported, suggested patients could discontinue oxaliplatin for a period of time while continuing 5-FU/leucovorin in the first-line setting without overall impact on survival. The follow up study, OPTIMOX-2, reported at ASCO 2007, attempted to extend these observations by randomizing patients to receive either 1) “maintenance therapy,” consisting of FOLFOX 7 x 6 cycles followed by LV, 5-FU 2 until tumor progression to baseline size, followed by reintroduction of FOLFOX 7 or 2) CFI therapy: FOLFOX 7 x 6 cycles and no maintenance until tumor progression followed by FOLFOX 7 reintroduction.[10] 

Two-hundred and two patients were randomized with similar baseline patient and disease characteristics.  Notably, 42% of patients in both arms had an ECOG PS 2.  This study, initially planned as a 600-patient Phase III trial was redesigned as a smaller trial which was able to detect a 20% difference in the two-year survival rate.  There was a strong trend towards improved overall survival in the maintenance arm vs. CFI, 26 months vs. 19 months (p=0.0549).  Overall responses were similar in both groups and median PFS was 36 weeks for maintenance vs. 29 weeks for CFI (p=0.08).  After reintroduction of FOLFOX 7, there was no significant difference in median PFS.  No grade 3 neuropathy was seen in initial FOLFOX 7 therapy in either arm, and remained in the 6-7% range after reintroduction of FOLFOX 7.  The authors concluded that a CFI is not an appropriate strategy after FOLFOX 7 first-line treatment. The excellent median overall survival and progression-free survival for these patients is striking and might be explained by the fact that only approximately 20% had prior adjuvant therapy, 76% had synchronous presentation of primary metastases, and half had only one metastatic site.  Nonetheless, these are outstanding results for chemotherapy-only treated patients.

Poor Performance Status Patients

Many patients in community practice present to their oncologist with a poor performance status.  Such patients are often excluded from Phase III trials.  Dr. Goldberg reported a summary of nine first-line mCRC studies with an intent to compare PS 0-1 patients to PS 2 patients for progression-free survival, overall survival and toxicity.[11] Approximately 10% of over 6,000 patients enrolled on these 9 trials had a PS of 2.  They were balanced for age, sex and study arm.  The progression free survival was worse for PS 2 patients, HR 1.52, p=<0.0001, as was overall survival HR 2.18, p=<0.0001, so PS 2 is prognostic for poor outcome. 

These investigators also evaluated whether more aggressive regimens benefit patients with PS 2. The response rate was higher with the more aggressive regimen in each study in both better and worse PS patients.  PFS showed similar results.  GI toxicity was greater for PS 2 patients and 60 day mortality was 12% compared to 3% for PS 0-1.  Surprisingly, there was no difference in neutropenia.  When looked at from the perspective of only trials that compared monotherapy to combination therapy, PFS was improved but not overall survival in PS 2 patients.  Infusional 5-FU based combination regimens were beneficial for all groups with a longer benefit in the PS 2 group.  These results suggest that patients with PS 2 with a diagnosis of metastatic disease should be considered for combination treatments as long as great attention to the increased toxicity is observed.

XELOX to FOLFOX Variations

Other studies also investigated XELOX to FOLFOX variations.  Ducreux performed a randomized, Phase III comparison study of XELOX to FOLFOX 6.[12] The primary endpoint was met with 42% response rate for XELOX vs. 46% for FOLFOX 6 which met statistical criteria for noninferiority.  PFS was nearly identical between arms at 8.8-9.3 months as was overall survival, 19.9-20.5 months.  XELOX caused less Grade 3/4 neuropathy, neutropenia and febrile neutropenia but more hand-foot syndrome.  

Second-line XELOX vs. FOLFOX in over 600 patients treated with first-line irinotecan containing regimens was reported by Rothenberg and colleagues.[13] Response rates for XELOX vs. FOLFOX were similar, 15.3% vs.12.4%, PFS 4.7 months vs. 4.0 months and OS 11.9 vs 12.6 months.  No unexpected toxicities were encountered.

Neoadjuvant Therapy

A major issue in the management of mCRC is how best to treat patients with resectable metastatic disease in the liver with curative intent.  While studies have clearly demonstrated that patients with R0 resections (no evidence of disease) have improved survival, the use of neoadjuvant versus adjuvant chemotherapy after resection versus both is not clearly established.  Therefore, the final results of an EORTC Study evaluating perioperative FOLFOX 4 chemotherapy and surgery for resectable liver metastasis was eagerly awaited and was the subject of a plenary presentation by Dr. Nordlinger.[14]  

Three hundred-sixty four patients were randomized to FOLFOX 4 x 6 cycles followed by surgery followed by six additional cycles of FOLFOX 4 versus surgery alone, the control arm.  Eligible patients had up to four lesions in the liver by CT scan and no extrahepatic disease.  Over three-quarters of patients in the chemotherapy arm received all planned preoperative cycles with good preservation of relative dose intensity.  Little severe toxicity was noted and no toxic deaths occurred.  Liver lesions shrunk on average from 45 mm to 30 mm, 43.9% of patients had a response to chemotherapy, 35.2% had stable disease and 6.6% had progressive disease.  Of 12 progressors, four underwent resection.

While more patients in the surgery only arm came to operation, (93.4% vs. 87.4%), the resection rate for both arms was nearly identical at 83.5% vs. 83.0%.  Postoperative complications were somewhat higher for the chemotherapy group.  Compliance with postoperative chemotherapy was less successful: 63% received treatment and only 44% of those received all planned six cycles. The primary endpoint, percentage difference in three year progression-free survival was 9.2% better for all resected patients, HR 0.73, p=0.025.  The authors concluded that the study approach of preoperative chemotherapy followed by postoperative chemotherapy in patients with resectable liver metastases should become a new standard.  However, the discussant, Dr. Petrelli, took the viewpoint that the overall benefit was modest and the toxicity of preoperative chemotherapy remains a concern.  He suggested that immediate liver resection followed by postoperative chemotherapy is an alternative approach that has also been tested and is proven.  He did recommend a trial comparing directly preoperative to postoperative chemotherapy. Such a trial is not yet underway in the US Intergroup system.  

Biomarkers        

The technology to detect circulating tumor cells (CTC) in metastatic human epithelial malignancies using immunomagnetic cell sorting technology is well established.  Dr. Meropol presented an analysis of CTC in mCRC.[15] His group looked at the number of CTCs at different time points compared to tumor response by imaging.  Peripheral blood was collected at baseline and at several times up to 20 weeks after treatment initiation with different lines of chemotherapy.  A training set of 109 patients and a validation set of 321 patients were generated.  Prior to the initiation of therapy, 53% of patients had no CTCs in 7.5 ml of blood while the other half were positive for 1 to >10 CTC.  A breakpoint of > 3 CTCs at 3-5 weeks after initiation of therapy separated response to treatment as indicated by imaging with a sensitivity of 27%, specificity of 93% and overall agreement of 78%.  

Baseline CTCs allowed differentiation of PFS and OS since patients who had greater > 3 CTCs at baseline had a worse prognosis regardless of therapy.  CTCs also provided prognostic information for both PFS and OS at the time of first follow up.  A decrease in CTCs at 3-5 weeks was associated with improved PFS in patients with >3 CTCs at baseline. These intriguing results suggest that we may soon be able to adopt CTC testing clinically in mCRC as clinicians have begun in metastatic breast cancer.  

Conclusions

The prospects for patients with metastatic colorectal cancer have improved substantially with the advent of new drugs and biological agents. Refinements in treatments allow simultaneous tumor shrinkage and improvement in overall quality of life with acceptable short and long term toxicity of the treatment. Increasingly patients with liver limited disease can be resected for cure, and the use of combined modality therapy in the treatment of mCRC is increasingly established. There remains more work to be done in refining the continuum of therapy with surgery, chemotherapy, and biological agents already available. A new generation of targeted therapeutics under investigation will hopefully continue to extend these encouraging results.

References:

1  Fuchs C, Marshall J, Mitchell E., et al.  Updated results of BICC-C study comparing first-line irinotecan/fluoropymidine combinations with or without celecoxib in mCRC: Updated efficacy data.  Proceedings from the American Society of Clinical Oncology Conference. Chicago, IL. 2007. Abstract # 4027. 

2  Saltz L, Clarke S, Diaz-Rubio E., et al.  Bevacizumab (Bev) in combination with XELOX or FOLFOX4: Updated efficacy results from XELOX-1/N016966, a randomized phase III trial in first-like metastatic colorectal cancer. Proceedings from the American Society of Clinical Oncology Conference. Chicago, IL. 2007. Abstract # 4028. 

3  Cassidy J, Clarke S, Diaz-Rubio E., et al.  XELOX compared to FOLFOX4: Survival and response results from XELOX-1/N016966, a randomized phase III trial of first-line treatment for patients with metastatic colorectal cancer (MCRC). Proceedings from the American Society of Clinical Oncology Conference. Chicago, IL. 2007. Abstract # 4030.  

4  Sobrero A, Young S, Balcewicz M., et al.  Phase IV study of first-line bevacizumab plus irinotecan and infusional 5-FU/LV in patients with metastatic colorectal cancer: AVIRI. Proceedings from the American Society of Clinical Oncology Conference. Chicago, IL. 2007. Abstract # 4068. 

5  Kopetz S, Glover K, Eng C., et al.  Phase II study of infusional 5-fluorouracil, leucovorin, and irinotecan (FOLFIRI) plus bevacizumab as first-line treatment for metastatic colorectal cancer. Proceedings from the American Society of Clinical Oncology Conference. Chicago, IL. 2007. Abstract # 4089. 

6  Schmiegel W, Reinacher-Shick A, Freier W., et al.  Comparable safety and response rate with bevacizumab in combination with capecitabine/oxaliplatin (CapOx/Bev) versus capecitabine/irinotecan (Capiri/Bev) in advanced CRC (mCRC): A randomized phase II study of the AIO GI tumor study group. Proceedings from the American Society of Clinical Oncology Conference. Chicago, IL. 2007. Abstract # 4034. 

7  Van Cutsem E, Nowacki M, Lang I., et al.   Randomized phase II study of irinotecan and 5-FU/FA with or without cetuximab in the first-line treatment of patients with metastatic colorectal cancer (mCRC): The CRYSTAL trial. Proceedings from the American Society of Clinical Oncology Conference. Chicago, IL. 2007. Abstract # 4000. 

8  Bokemeyer C., Bondarenko I, Makhson A., et al.  Cetuximab plus 5-FU/FA/oxaliplatin (FOLFOX-4) versus FOLFOX-4 in the first-line treatment of metastatic colorectal cancer (mCRC): OPUS, a randomized phase II study. Proceedings from the American Society of Clinical Oncology Conference. Chicago, IL. 2007. Abstract # 4000. 

9   Punt C, Koopman M, Douma J., et al.  Sequential compared to combination chemotherapy with capecitabine, irinotecan, and oxaliplatin in advanced colorectal cancer (ACC):  A Dutch Colorectal Cancer Group (DCCCG) phase III study. Proceedings from the American Society of Clinical Oncology Conference. Chicago, IL. 2007. Abstract # 4012. 

10  Maindrault-Goebel F, Lledo G, Chibaudel B., et al.  Final results of OPTIMOX2, a large randomized phase II study of maintenance therapy or chemotherapy-free intervals (CFI) after FOLFOX in patients with metastatic colorectal cancer (MRC): A GERCOR study. Proceedings from the American Society of Clinical Oncology Conference. Chicago, IL. 2007. Abstract # 4013. 

11  Goldberg R, Köhne C, Seymour M., et al.  A pooled safety and efficacy analysis examining the effect of performance status (PS) on outcomes in nine first-line treatment (rx) trials (cts) of 6,286 patients (pts) with metastatic colorectal cancer (MCRC). . Proceedings from the American Society of Clinical Oncology Conference. Chicago, IL. 2007. Abstract # 4011. 

12  Ducreux M, Bennouna J, Hebbar M., et al.  Efficacy and safety findings from a randomized phase III study of capecitabine (X) + oxaliplatin (O) (XELOX) vs. infusional 5-FU/LV = O (FOLFOX-6) FOR METASTATIC COLORECTAL CANCER (mcrc).  Proceedings from the American Society of Clinical Oncology Conference. Chicago, IL. 2007. Abstract # 4029. 

13  Rothenberg M, Navarro M, Butts C., et al.  Phase III trial of capecitabine + oxaliplatin (XELOX) vs. 5-fluorouracil (5-FU), leucovorin (LV), and oxaliplatin (FOLFOX4) as 2nd-line treatment for patients with metastatic colorectal cancer (MCRCAbstract Proceedings from the American Society of Clinical Oncology Conference. Chicago, IL. 2007. Abstract # 4031. 

14.  Nordlinger B, Sorbye H, Collette L., et al.  Final results of the EORTC Intergroup randomized phase III study 40983 (EPOC) evaluating the benefit of peri-operative FOLFOX4 chemotherapy for patients with potentially resectable colorectal cancer liver metastases.  Proceedings from the American Society of Clinical Oncology Conference. Chicago, IL. 2007. Abstract LBA5. 

15  Meropol N, Cohen S, Iannotti N., et al.  Circulating tumor cells (CTC) predict progression free (PFS) and overall survival (OS) in patients with metastatic colorectal cancer. .  Proceedings from the American Society of Clinical Oncology Conference. Chicago, IL. 2007. Abstract # 4010.