A funny thing happened on the way to the way to the annual ISGIO meeting in Washington, DC, September 13–15, 2012. After surviving various legitimate medical misfortunes of some consequence over the past few years—and completing a number of trips to India and Southeast Asia uncomplicated by any exotic infirmities—I ironically managed to acquire an illness in North America more typically associated with less salubrious locales. Waylaid by this mischievous malady, I was unable to attend this year's annual symposium. This frustrating development upset me considerably, as I missed out on the educational opportunities and camaraderie associated with ISGIO that I have come to value immensely and so look forward to every fall. Moreover, my official duties were few this year, and I would have been free to afford myself a fuller measure of the benefits ISGIO has to offer regular attendees and would have had plenty of time to engage in the collegiality I so enjoy. I was asked to deliver a brief presentation on the state of the GI Intergroup at ISGIO. Since I was unable to be present at the meeting, I thought I would take this opportunity to relate some of those remarks.
As co-chair of the GI Steering Committee with Joel Tepper and Neal Meropol, I find my job to be a seeming continuation of the first position I held after finishing fellowship, when I worked at the Cancer Therapy Evaluation Program (CTEP) of the NCI. In that position, I helped manage portfolios that included cooperative groups such as the Eastern Cooperative Oncology Group (ECOG), the Gynecologic Oncology Group (GOG), and contract groups, such as the Gastrointestinal Study Group (GITSG). “Plus ça change, plus c'est la même chose.” Having said that, I must note that many changes have taken place in protocol generation and approval over the past three decades.
In earlier times, a relatively small cadre of individuals within each cooperative or contract group generated protocols and an equally small CTEP body almost always granted approvals. Those who now participate in the current Intergroup know that the system has become more complicated, while, it is hoped, becoming more inclusive and productive. The point has been made that the current cooperative group system arose from the need to perform large, multi-institutional trials in relatively common (then thought to be homogeneous) malignancies, such as leukemia, breast cancer, and other solid tumors. Tools were generally limited to a small selection of standard chemotherapeutics, and tumor biology for patient or drug selection was in its infancy.
Most of us are aware that many of these characteristics of early cooperative group clinical trials have greatly changed. Even after the merger of the groups into just four adult entities, there are still multiple concept sources, even without the Cancer Centers, P-01s, and R-01s. Common solid tumors, such as colon, breast, and lung cancer, as well as hematologic malignancies, are no longer homogeneous or monolithic diseases amenable to simple protocol study designs. Newer biologic agents call for the development of predictive markers and innovative study designs to avoid both over- and undertreatment. For gastrointestinal oncology, two straightforward instances of this evolution have been the identification of KRAS and HER2 in colon and gastric cancers, respectively. At a time when obtaining financial resources has become more competitive, it has become increasingly desirable to have clinical trials integrated with real-time monitoring, pathology review, and treatment planning; translational research projects; and health-related quality-of-life indices.
What other complexities have arisen? In my time at CTEP, virtually all investigational drugs were sponsored by the NCI Investigational Drug Branch, so that issues of drug availability were largely dealt with internally. Currently, the vast majority of drug supply comes directly from pharmaceutical companies who sponsor their own pivotal trials and have their own priorities for drug development. A number of studies, for example, that passed muster within a cooperative group and various stages within the GI Intergroup became moot when drug development priorities within a pharmaceutical firm changed. Additionally, there is much research ongoing internationally. In some instances, these efforts are integral to pharmaceutical industry projects and add to the focus and nimbleness of such trials. As smaller subsets of patients are identified for treatment, particularly with targeted biologics, more patients will have to be screened even beyond the borders of North America.
North American cooperation with Western Europe, for example, seems obvious, but is fraught with issues related to official agreements, drug distribution, and differences of opinion about study questions and trial design among proposed partners. All too often, such discussions do not begin to take place until a proposing entity has a rather mature study, requiring buy-in at a later date by other groups. An obvious case in point may be the development of a gastric adjuvant trial with trastuzumab. The preclinical and clinical data are strong: there is both a predictive marker, a strong signal for activity from the ToGA trial in advanced disease, and evidence from MAGIC, Japanese, and Korean trials that adjuvant or perioperative therapy may be as effective as chemoradiotherapy derived from INT-0116. Meanwhile, a Dutch group is studying the necessity of postoperative radiotherapy with perioperative chemotherapy in the CRITICS trial, and so a portfolio of adjuvant options is developing: the evaluation of trastuzumab seems worthy. HER2 expression appropriate for prediction of trastuzumab activity is uncommon, occurring in about 15% of gastric cancers in the West. Of course, the hypothesis assumes that micrometastases have the same biologic pathways as advanced disease, which was not apparently seen with EGFR signaling in colon cancer with cetuximab, but, it is hoped, will replicate the translation of trastuzumab activity in breast cancer to the adjuvant setting. To screen and accrue adequate numbers of patients for a phase III trial will require at least the participation of all North American groups and, it is hoped, that the recruitment will attract international participation from Western Europe and Australasia.
This potential study illustrates some of the issues facing complicated, multimodal, biologically driven clinical trials of the current era. In the otherwise collaborative environment of GI cancer research, particularly within the U.S. Cooperative Groups, ideas for studies may seem self-evident to many, but development requires leadership from a small cadre of responsible investigators, with the commitment of many. In the current environment, ideas typically arise from discussions within disease-oriented Cooperative Group committees. After vetting, ideas are brought forward to the task forces of the GI Intergroup, who then ultimately present protocol concepts to the GISC.
Over the years, approximately 50% of such concepts have been approved. None approved by GISC have failed to be approved and funded by CTEP, although some studies have been terminated early because of underaccrual. As one of the oldest Intergroups, the GI Intergroup, has weathered many storms and learned to navigate the winding waters leading to clinical trials development in the modern era. Twice-yearly, face-to-face GISC meetings are simply not sufficient to move protocols forward, incorporate the cumulative wisdom of a nonpareil group of clinical investigators, and remain competitive with non-U.S. groups and the pharmaceutical industry.
Monthly Task Force calls take place, with CTEP and one or more of the co-chairs of the GISC participating, and plans are under way for more continuous input into concept development to enhance the likelihood that randomized phase II and III designs presented to GISC will be approved. Evaluation at the Task Force level of the most commonly occurring reasons for disapproval of a concept will help to reduce this likelihood: accrual concerns, extent of preclinical data, lack of a biologically driven question, and a few other matters form the most predictable and presumably addressable issues. And, as the scenario for an HER2-driven gastric adjuvant study suggests, early international and pharmaceutical collaboration and cooperation are imperative. The days of single cooperative groups performing large chemotherapy trials in unselected patients are very much part of my history, but not of the future of research in gastrointestinal cancer.