Originally published as JCO Early Release 10.1200/JCO.2008.20.6284 on March 9 2009

This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Schmoll, H.-J. Articles by Arnold, D. Search for Related Content PubMed PubMed Citation Articles by Schmoll, H.-J. Articles by Arnold, D. Related Articles Related Article Social Bookmarking                            What's this?

When Wishful Thinking Leads to a Misty-Eyed Appraisal: The Story of the Adjuvant Colon Cancer Trials With Edrecolomab

Department of Internal Medicine IV, Martin Luther University Halle-Wittenberg, Halle, Germany

Colorectal cancer is the most frequently occurring cancer in the Western world, with a worldwide incidence of about 1 million, and with 70% to 80% of patients presenting initially with potentially curable stage I to III disease, according to International Union Against Cancer criteria of tumor staging. Today, the 5-year survival rates of patients with high-risk stage II and III colon cancer treated with adjuvant fluorouracil (FU) plus oxaliplatin chemotherapy and patients with stage II to III rectal cancer treated with neoadjuvant chemoradiotherapy are 78.5% and 67%, respectively, whereas in the 1980s, the survival rate of patients with stage III colon and rectal cancer undergoing surgery alone was only 45% to 55%.^1–3 Therefore, 20 years ago, any potentially effective adjuvant treatment modality was highly valued, prompting large phase II to III trials, such as those involving FU plus levamisole and leucovorin. However, because treatment with chemotherapy is associated with toxicity, and a majority of patients treated receive no individual benefit, immunotherapeutic approaches were regarded as attractive options.

The discovery of the transmembrane glycoprotein epithelial cell adhesion molecule (EpCAM), which is present on many human epithelia and overexpressed in many malignancies, including colorectal cancer, led to the development of edrecolomab, an EpCAM murine monoclonal antibody directed against this epitope. Preclinical data demonstrated the ability of edrecolomab to induce antibody-dependent cell-mediated cytotoxicity, and early clinical data from Herlyn et al⁴ at the Wistar Institute (Philadelphia, PA) demonstrated some antitumor efficacy, even in patients with advanced disease, in addition to immunologic effects like induction of a secondary antibody as an internal imaging effect. This antibody, therefore, went on to be used in a small phase III trial⁵ in patients with resected stage III colorectal cancer, before efficacy in patients with advanced disease was formally demonstrated.

This first study by Riethmüller et al⁵ was initiated in 1985 and accrued until 1990, in parallel with large trials by the North Central Cancer Treatment Group, National Surgical Adjuvant Breast and Bowel Project, and other groups addressing the role of adjuvant chemotherapy with FU plus leucovorin.^2,3 In contrast to these large trials, the first edrecolomab study⁵ accrued only 189 randomly assigned and 166 eligible patients (88% of the intent-to-treat population). After a median follow-up of 5 years in 1995⁵ and 7 years in 1998,⁶ the authors concluded there was significant superior relapse-free and overall survival favoring edrecolomab, with a survival benefit of the same magnitude as that observed with FU plus leucovorin. These results prompted approval of edrecolomab for routine use in adjuvant therapy in Germany, and were heralded as marking the beginning of a new age of immunotherapy, at least in colon cancer. Subsequently, to confirm the results of this pivotal trial, four large prospective randomized trials were initiated in patients with stage II and III colon cancer.^7–10

In this issue of Journal of Clinical Oncology (JCO), the final results of the study by Fields et al¹⁰ have been published, confirming the lack of efficacy of edrecolomab reported in the other three trials, which compared edrecolomab alone with no treatment in stage II disease (Table 1),^7,8 and edrecolomab alone or in combination with FU plus leucovorin with chemotherapy in stage III colon cancer (Table 2).⁹ The single-agent activity of edrecolomab reported in the original study by Riethmüller et al^5,6 was not reproduced in either stage II or stage III disease. In fact, results in the edrecolomab single-agent arm of the trial⁹ of patients with stage III disease were significantly inferior to those seen with FU plus leucovorin, and correlated well with the results of surgery alone in the early adjuvant FU plus leucovorin trials.^2,3 The combination of edrecolomab with FU plus leucovorin also did not improve outcomes. In patients with stage II disease, single-agent edrecolomab showed no efficacy compared with surgery alone. These four confirmatory trials^7–10 have demonstrated convincingly the complete lack of efficacy of edrecolomab alone or in combination with chemotherapy in stage II and III colon cancer (Tables 1 and 2).

How could such a small—albeit randomized—study have led to such a large investment? More importantly, how could this have been prevented? When Herlyn et al⁴ investigated the immunologic effects of edrecolomab in a small series of patients with advanced colorectal cancer, tumor regression and longer survival were seen in some patients, in addition to the expected immunologic effects. Unfortunately, this cohort also included patients receiving FU chemotherapy in combination with edrecolomab, so the efficacy of the antibody may have been overrated. However, after this observation—and in light of the frequently expressed hypothesis that antibody-based immunotherapy might be most active in minimal residual disease and less effective in more bulky, measurable tumor masses (which was supported by the observation that edrecolomab only worked in nude mice xenografts if it was injected at the time of tumor cell inoculation¹¹)—it was concluded that edrecolomab might be active in the postoperative adjuvant treatment setting. This hope for the efficacy of immunotherapy in minimal residual disease only also led to the idea that such a treatment would probably not be active in advanced disease, and therefore did not require demonstration of efficacy in advanced disease before testing in the adjuvant setting.

This conclusion, although clinically unproven, might have been understandable in an era of no alternative adjuvant therapy. However, despite the fact that this trial⁵ was constructed in accordance with research practices common at that time, it is not in retrospect understandable why the so-called positive results mandated additional investigation of this approach. Innovators are by nature advocates of their therapies, but in this case, the scientific community or sponsor must have grown somewhat misty eyed with enthusiasm and hope, and been misled by so-called significant data.¹² A more critical analysis of the investigated patient cohort and the data from the original trial^5,6 probably should have raised a higher level of caution and reluctance. A couple of severe drawbacks in the methodology and analyses of the study, and inconclusive data in both reports, should have raised the need for just one medium-sized confirmatory trial before additional large trials were established.

First, 12% of patients in the study⁵ were determined to be ineligible, leaving only 166 evaluable patients of the intent-to-treat population of 189 patients; analyses were based primarily on the evaluable patient population. Also, the study included both patients with colon cancer and patients with rectal cancer (42%), two groups with different local relapse rates, particularly dependent on whether adjuvant chemoradiation is applied. This mixture of patients with colon cancer and patients with rectal cancer, who may or may not have received adjuvant chemoradiation, introduced potentially relevant imbalances and biases. Furthermore, because most patients in the antibody arm did not receive adjuvant chemoradiation, there was a nearly two-fold local relapse rate as high as 25% in the edrecolomab arm, compared with a rate of 15% in the control arm. This rate was high for a study in which 58% of patients had colon cancer, a disease in which local relapse should be a rare exception.

In addition, prognostic factors were not evenly distributed between the arms, with more patients with early stage T1 and G1-2 disease in the experimental arm; this was masked in the original report by the use of the wrong denominator for calculation of their distribution. Also, randomization was performed according to the methodology proposed by Zelen,¹² with informed consent after randomization, giving patients the option to refuse the allocated treatment. This happened with seven patients. Even in 1985, this methodology was not commonly regarded as adequate due to possible introduction of a randomization bias. Furthermore, in the first report,⁵ the results were presented in a way suggesting clear statistical significant improvement in relapse-free and overall survival, although this was not the case according to standard criteria.

P values for recurrence-free and overall survival as well as for distant metastases and local relapse-free survival as presented in the figures and abstract were not log rank, but instead "adjusted for imbalances in prognostic factors" with a Cox model.⁵ The log-rank analysis of these survival curves, as reported in the text of the article (in contrast to the abstract and figures), showed a nonsignificant survival benefit (P = .051). This is consistent with the large and widely overlapping confidence intervals. In the final analysis,⁶ there was again no significant difference in recurrence-free survival (P = .07), but the overall survival results clearly favored edrecolomab (P = .01). However, it is unclear whether the tests for significance were performed with two-sided calculations.

All these issues could have introduced major bias into the validity of the data, particularly given the small number of randomly assigned patients, the high ineligibility rate, and the mixture of and imbalance in relevant prognostic subgroups. All these drawbacks could have been identified in 1994, before the start of the confirmatory trial program.

From today's point of view, the era of monoclonal antibodies was ushered into clinical practice in oncology by edrecolomab, and despite its ineffectiveness, it helped to pave the way for additional investigation of monoclonal antibodies in colorectal cancer. This stimulated the search for better molecular targets directed not just against epitopes, which were thought to be more or less nonfunctional, with antitumor effects mostly limited to induction of antibody-dependent cell-mediated cytotoxicity. It also opened the door for agents directed against functional domains like the endothelial growth factor receptor family and the consecutive inhibition of the dependent pathways. Additional developments have addressed antiangiogenesis and interaction between tumor and stroma. Remarkably, metastatic colorectal cancer again served as the model for development of these strategies.

A better understanding of antigen-related immunology has also contributed to the concept that even so-called nonfunctional antigens may be involved in cellular signaling (eg, antibodies against CD20, which have been shown to induce direct proapoptotic effects). Although interest in edrecolomab has been suspended, a new chapter in the book on EpCAM has been opened by the discovery that EpCAM is involved in many mechanisms of tumor proliferation. A genome-wide expression analysis has identified the EpCAM receptor as part of a large network of regulators for cell cycle, cell death, and cellular growth. Also, EpCAM is a ligand to the human leukocyte immunoglobulin–like receptor 1, which provides a link to immune regulation via natural killer cells, lymphocytes, and dendritic cells. Therefore, upregulation of EpCAM in colon carcinomas may lead to direct downregulation of local immunity, and thus support escape from immune surveillance. In this concept, an anti-EpCAM antibody may contribute as a stimulant for normalization of immune surveillance and facilitate a clearance of the tumor. If completely understood, all these effects—rather than the direct cytotoxic effects of the anti-EpCAM antibodies—could play important roles in a new era of immunotherapy, potentially in combination with other molecular agents and in resulting immune modulation. In this setting, MT201—a novel, fully humanized immunoglobulin G1 monoclonal antibody—has entered into clinical trials for a number of tumors, including a trial of adjuvant treatment of resected colorectal cancer metastases, despite the lack of benefit shown in advanced measurable disease.¹⁴

Immunotherapy and development of molecular agents against human malignancies are rapidly developing fields, and improvements have been regarded with fascination and enthusiasm. However, the rush of innovations observed at the bench and in preclinical models must be translated carefully for the clinical setting. One might ask whether the edrecolomab scenario could happen again. One important answer is that a higher awareness of statistical needs and limitations—along with a more stringent methodology with strict regulations from authorities, ethical committees, and clinical trial steering and data monitoring boards—may be preventive. However, there are currently many planned and ongoing trials in which financial limitations have led to the design of underpowered comparisons with seemingly constructed end points of efficacy, mostly in the methodology of randomized phase II trials. Therefore, the edrecolomab experience may serve as a case in point for the need for excellence in clinical research, with reasonable designs, evaluations, and conclusions without misty eyes.

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Although all authors completed the disclosure declaration, the following author(s) indicated a financial or other interest that is relevant to the subject matter under consideration in this article. Certain relationships marked with a "U" are those for which no compensation was received; those relationships marked with a "C" were compensated. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.

Employment or Leadership Position: None Consultant or Advisory Role: Hans-Joachim Schmoll, Roche (C), Sanofi-aventis (C), Merck Sharp & Dohme (C); Dirk Arnold, Roche (C), Pfizer (C), Merck Sharp & Dohme (C) Stock Ownership: None Honoraria: Hans-Joachim Schmoll, Roche, Sanofi-aventis, Merck Sharp & Dohme; Dirk Arnold, Roche, Pfizer, Merck Sharp & Dohme Research Funding: None Expert Testimony: None Other Remuneration: None

AUTHOR CONTRIBUTIONS

Conception and design: Hans-Joachim Schmoll, Dirk Arnold

Collection and assembly of data: Hans-Joachim Schmoll

Data analysis and interpretation: Hans-Joachim Schmoll, Dirk Arnold

Manuscript writing: Hans-Joachim Schmoll, Dirk Arnold

Final approval of manuscript: Hans-Joachim Schmoll, Dirk Arnold

REFERENCES

1. Meyerhardt JA, Mayer RJ: Systemic therapy for colorectal cancer. N Engl Med 352:476–487, 2005.[Free Full Text]

2. Moertel CG, Fleming TR, Macdonald JS, et al: Fluorouracil plus levamisole as effective adjuvant therapy after resection of stage III colon carcinoma: A final report. Ann Intern Med 122:321–326, 1995.[Abstract/Free Full Text]

3. Efficacy of adjuvant fluorouracil and folinic acid in colon cancer. International Multicentre Pooled Analysis of Colon Cancer trials (IMPACT) investigators. Lancet 345:939–944, 1995.[CrossRef][Medline]

4. Herlyn D, Wettendorff M, Schmoll E, et al: Anti-idiotype immunization of cancer patients: Modulation of the immune response. Proc Natl Acad Sci U S A 84:8055–8059, 1987.[Abstract/Free Full Text]

5. Riethmüller G, Schneider-Gädicke E, Schlimok G, et al: Randomised trial of monoclonal antibody for adjuvant therapy of resected Dukes' C colorectal carcinoma: German Cancer Aid 17-1A Study Group. Lancet 343:1177–1183, 1994.[CrossRef][Medline]

6. Riethmüller G, Holz E, Schlimok G, et al: Monoclonal antibody therapy for resected Dukes' C colorectal cancer: Seven-year outcome of a multicenter randomized trial. J Clin Oncol 16:1788–1794, 1998.[Abstract]

7. Hartung G, Hofheinz RD, Dencausse Y, et al: Adjuvant therapy with edrecolomab versus observation in stage II colon cancer: A multicenter randomized phase III study. Onkologie 28:347–350, 2005.[CrossRef][Medline]

8. Colacchio TA, Niedzwiecki D, Compton C, et al: Phase III trial of adjuvant immunotherapy with MOAb 17-1 following resection of stage II adenocarcinoma of the colon (CALGB 9581). J Clin Oncol 22:250s; 2004 (suppl) abstr 3522.

9. Punt CJ, Nagy A, Douillard YJ, et al: Edrecolomab alone or in combination with fluorouracil and folinic acid in the adjuvant treatment of stage III colon cancer: A randomised study. Lancet 360:671–667, 2002.[CrossRef][Medline]

10. Fields A, Keller A, Schwartzberg L, et al: Adjuvant therapy with the monoclonal antibody edrecolomab plus fluorouracil-based therapy does not improve overall survival of patients with stage III colon cancer. J Clin Oncol 27:1941–1947, 2009.[Abstract/Free Full Text]

11. Herlyn D, Steplewski Z, Herlyn M, et al: Inhibition of growth of colorectal carcinoma in nude mice by monoclonal antibody. Cancer Res 40:717–721, 1980.[Abstract/Free Full Text]

12. Ioannidis JP: Contradicted and initially stronger effects in highly cited clinical research. JAMA 294:218–228, 2005.[Abstract/Free Full Text]

13. Zelen M: A new design for randomized clinical trials. N Engl J Med 300:1242–1245, 1979.[Abstract]

14. Micromet Key Events for 2009. http://www.bio-medicine.org/medicine-technoloy-1/Micromet-Key-Events-for-2009-3654-2/.

CiteULike    Complore    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

Related Article

• Adjuvant Therapy With the Monoclonal Antibody Edrecolomab Plus Fluorouracil-Based Therapy Does Not Improve Overall Survival of Patients With Stage III Colon Cancer
  Anthony L.A. Fields, Alan Keller, Lee Schwartzberg, Stephen Bernard, Carl Kardinal, Alan Cohen, Joe Schulz, Peter Eisenberg, John Forster, and Paul Wissel
  JCO 2009 27: 1941-1947 [Abstract] [Full Text]

This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Schmoll, H.-J. Articles by Arnold, D. Search for Related Content PubMed PubMed Citation Articles by Schmoll, H.-J. Articles by Arnold, D. Related Articles Related Article Social Bookmarking                            What's this?