Gastrointestinal Cancers: May 25, 2010

New Research and Emerging Trends from the World Congress on Gastrointestinal Cancer

The 10th Annual Meeting of the World Congress on Gastrointestinal Cancer was held in Barcelona, Spain, in June. The Congress has grown into a key scientific program that covers malignancies affecting every component of the gastrointestinal tract and all facets of patient care, including screening, diagnosis, and management options for common and uncommon tumors.

As was pointed out at the meeting, the number of deaths from colorectal cancer has been dropping in recent years. Polyps are being found by screening and removed before they can develop into cancers. Screening is also allowing more colorectal cancers to be found earlier, when the disease is easier to cure. In addition, treatment for colorectal cancer has improved over the past decade, allowing for more effective options for people with this diagnosis.

Presentation of cutting-edge research has become one of the highlights of the World Congress on Gastrointestinal Cancer, offering opportunities for researchers to interact with colleagues and share their research with the community. Some of the key sessions are summarized in this issue.

Colorectal Cancer

KRAS Gene Status Predicts Outcome of First-line Treatment in Patients with Metastatic Colorectal Cancer

Eric Van Cutsem, MD, PhD, University Hospital Gasthuisberg, Leuven, Belgium, presented results from a subanalysis of the CRYSTAL study. The data show that patients with metastatic colorectal cancer are more likely to respond to first-line therapy of cetuximab (Erbitux) in combination with standard chemotherapy if they have wild-type KRAS tumors.

In 2007, the randomized phase III CRYSTAL trial showed that some patients with metastatic disease had longer progression-free survival (PFS) from FOLFIRI chemotherapy plus cetuximab than from FOLFIRI alone. Not all patients benefited, however, prompting the researchers to conduct a subanalysis to determine the influence of KRAS status on the outcome of initial combination therapy. The researchers analyzed archived tumor material from 578 of the 1,198 randomized patients enrolled in the CRYSTAL study. Of these, 540 samples were found to be KRAS-evaluable— 348 with normal KRAS status and the others with KRAS mutations— and were evaluated using PFS as the primary endpoint.

Of the patients with wild-type KRAS, 59% responded to chemotherapy with cetuximab, while 43% responded to chemotherapy alone. The mean PFS duration was 9.9 months for patients receiving combination therapy and 8.7 months for those getting FOLFIRI treatment alone.

The type of therapy given to patients with mutated KRAS status did not make much difference in the overall response rate (ORR) and PFS. The ORR for patients receiving combination therapy was 36%, compared with 40% for standard therapy. Initial combination treatment resulted in a mean PFS of 7.6 months, while chemotherapy led to 8.1 months.

The results indicate that KRAS status could be an important predictor of the effectiveness of metastatic colorectal cancer treatment. “KRAS is the first molecular marker for targeted therapy in combination with standard chemotherapy as a first-line treatment for metastatic colorectal cancer,” said Dr. Van Cutsem. “If we know in advance that a patient has a KRAS mutation, then we know we don’t have to treat the patient [with these agents].”

Bevacizumab Shows Benefit in Metastatic Colorectal Cancer Independent of KRAS Mutation Status

A team of researchers from Genentech, Inc., San Francisco, California, and Duke University, Durham, North Carolina, conducted a placebo-controlled phase III trial examining the benefits of bevacizumab (Avastin) in patients with metastatic colorectal cancer. Their findings, presented by Herbert Hurwitz, MD, Duke University Medical Center, indicate that the clinical benefits of the drug are similar in patients with KRAS wild-type tumors and mutant KRAS tumors.

Studies have shown that KRAS mutations, present in an estimated 30% to 40% of colorectal tumors, make it unlikely for patients to benefit from drugs such as cetuximab (Erbitux) that inhibit the activity of the epidermal growth factor receptor protein, which is often overactive in colorectal cancer. Bevacizumab, in contrast, works by destroying blood vessels that the tumors need to grow.

Dr. Hurwitz and his team of researchers microdissected tumor samples from 230 patients who were enrolled in a large, placebo-controlled randomized phase III study of irinotecan/fluorouracil/leucovorin (IFL) given with or without bevacizumab. Of this population, 129 patients received combination treatment while 101 were given placebo. The tumor samples were subjected to DNA sequence analysis for KRAS mutation, and researchers retrospectively measured progression-free survival (PFS), overall survival (OS), and objective response rate.

Among the 152 patients with wild-type KRAS tumors, median PFS was 13.5 months in the group getting IFL bevacizumab, compared with 7.4 months for the group getting IFL plus placebo, corresponding to a hazard ratio (HR) of 0.44. For the 78 patients with mutant KRAS, the IFL bevacizumab group had median PFS of 9.3 months, while those in the placebo group had a median PFS of 5.5 months (HR = 0.41).

In the wild-type KRAS population, the group getting IFL bevacizumab had median OS of 27.7 months, while the placebo group had median OS of 17.6 months, corresponding to a HR for death of 0.58. Among those with mutant KRAS, IFL bevacizumab led to a median survival of 19.9 months, as compared with 13.6 months for the placebo group (HR = 0.69).

Sixty percent of patients in the wild-type group responded to combination therapy compared with 37.3% to IFL. In the mutant KRAS group, combination therapy led to a response rate of 43.2% compared with 41.2% for IFL alone. “The response rate (60%), pro- gression-free survival (13.5 mo), and overall survival (28 mo) for patients whose tumors are KRAS wild type who were treated with bevacizumab are among the best reported in any colorectal cancer study to date,” said Dr. Hurwitz.

“Based on the available data, the decision to use bevacizumab can be made without KRAS testing,” Dr. Hurwitz added. “The relative and absolute benefits in terms of response rate, progression-free survival, and overall survival all suggest that a bevacizumab-containing regimen should remain the preferred default first-line approach to metastatic colorectal cancer. We wouldn’t recommend that patients be tested for KRAS mutations to predict a response to bevacizumab-based regimens.”

Mutant KRAS Predicts Response to Cetuximab in Advanced Colorectal Cancer; Prognostic Significance Unclear

The results of another study on the role of KRAS mutation status as a biomarker for the response to cetuximab (Erbitux) in colorectal cancer showed that the mutations can predict response to the medicine but they may not have a prognostic role. These results of NCIC CTG CO.17, a phase III trial of cetuximab versus best supportive care, were presented by Christos Karapetis, MD, University of Sydney, Sydney, Australia.

Dr. Karapetis and his colleagues collected tumor samples from 394 patients (69% of the total study population) with advanced colorectal cancer. Of these, 164 patients had mutant KRAS. Patients received either best supportive care or cetuximab monotherapy plus best supportive care. Cetuximab works by inhibiting the epidermal growth factor receptor (EGFR), which is often overactive in colorectal cancer.

Among the group of patients with mutant KRAS, combination therapy led to a median progressionfree survival (PFS) of 1.8 months. Median duration of overall survival (OS) was 4.6 months with cetuximab and 4.5 months with best supportive care (hazard ratio [HR] for death = 0.98).

In the wild-type KRAS population, the cetuximab- treated group had median PFS of 3.8 months, double the median PFS of 1.9 months for the group getting best supportive care (HR = 0.40). Median OS was also nearly double for the patients with wildtype KRAS who were receiving cetuximab versus those with best supportive care: 9.5 months and 4.8 months respectively, corresponding to a HR of 0.55.

These results show that there is an almost doubling of median OS and PFS in patients with wildtype KRAS tumors, while there is no benefit seen in patients with KRAS mutations. The researchers concluded that “in this population, KRAS mutation status is a strong predictive biomarker and KRAS mutation analysis may now be considered a new standard of care in the selection of patients for EGFR-targeted therapy.”

Pancreatic Cancer

Correlation of Diastolic Blood Pressure with Survival in Phase II Study of Gemcitabine-Axitinib in Advanced Pancreatic Cancer

Jean-Philippe Spano, MD, PhD, Hopital de la Pitie Salpetriere, Paris, France, presented the results of a preliminary analysis of a randomized phase II clinical trial of axitinib (AG- 013736) plus standard of care therapy gemcitabine in patients with advanced pancreatic cancer. The researchers investigated the relationship between axitinib concentration, survival, and diastolic blood pressure. Their analysis showed that the combination therapy seemed to prolong survival in patients with diastolic blood pressure of at least 90 mm Hg.

Axitinib is a potent and selective oral inhibitor of vascular endothelial growth factor (VEGF) receptors 1, 2, and 3. Inhibiting VEGF binding blocks the growth of blood vessels and starves tumors of nutrients they need to proliferate. Last year, Dr. Spano and his colleagues presented interim results of their clinical trial showing that the axitinib-gemcitabine combination therapy demonstrated a trend toward prolonged overall survival (OS). The results have warranted a randomized, doubleblind phase III trial.

In the phase II study, 103 previously untreated patients with advanced pancreatic cancer were randomized to receive either gemcitabine (GEM) alone or a combination of axitinib and gemcitabine (GA). The patients received 1,000 mg/m2 of GEM over 30 minutes on days 1, 8, and 15 every four weeks. Axitinib 5 mg was administered orally to 69 patients in the GA arm. The researchers took full pharmacokinetic profiles on cycle 1, day 1, for GEM; cycle 1, day 14, for steady-state axitinib alone; and cycle 1, day 15, for steady-state axitinib plus GEM.

Dr. Spano and his colleagues found that the pharmacokinetic parameters of GEM were similar in the presence of axitinib. The plasma AUC (area under concentration time curve) for GA was 11348 ng.h/mL while for GEM it was 12840 ng.h/mL. The AUC for the GEM metabolite dFdU (2’,2’-difluorodeoxyuridine) was 184511 versus 195051 ng.h/mL for GA and GEM, respectively.

Among the patients in the GA arm, the median overall survival and 1-year survival were 12.2 months and 51%, respectively, for the 38 patients with diastolic blood pressure (dBP) ≥ 90 mm Hg. For the remaining 31 patients, median OS and 1-year survival were 5.2 months and 19%, respectively. Median OS and 1-year survival were 5.6 months and 23% for patients receiving GEM.

“Preliminary analysis of patients with advanced pancreatic cancer treated with GA indicates that patients with dBP ≥90 mm Hg appear to be associated with longer survival,” the researchers concluded.

Gastrointestinal Stromal Tumor

Survival and Safety of Sunitinib in Patients with Imatinib-Resistant/Intolerant GIST

Piotr Rutkowski, MD, PhD, of the M. Sklodowska- Curie Memorial Cancer Center and Institute of Oncology, Warsaw, Poland, presented results of a detailed analysis of survival and safety in a worldwide treatment-use trial of sunitinib (Sutent).

The primary objective of this research was to grant access to sunitinib to patients who were ineligible for clinical trials because of entry criteria or who were from a country where sunitinib is pending regulatory approval and no GIST trials were available. Assessment parameters included safety and tolerability of the drug, tumor measurements, time to progression (TTP) and overall survival (OS).

As of December 2007, there were 1,126 patients enrolled in the study with a median time on treatment of 30 weeks. Fifty-nine percent of patients (661) underwent dose interruptions, and 42% (465 patients) had dose reductions. The median follow-up time was 51 weeks.

Of the intent-to-treat population, comprising 1,117 patients, 12 completed treatment. There were 906 discontinuations, mostly because of disease progression (46%) and adverse events (AEs; 19%). In the ITT population, Dr. Rutkowski found median TTP to be 41 weeks and median OS of 75 weeks. At data cut-off, 564 patients (50%) were still alive.

The most common treatment-related non-hematologic AEs (all grades) included fatigue (42%), diarrhea (39%), nausea (28%), and hand— foot syndrome (27%). Treatment-related hematologic AEs were thrombocytopenia (19%), neutropenia (18%), and anemia (14%). Ten percent of patients suffered treatment-related hypothyroidism (1% grade 3/4). Grade 3/4 lab abnormalities associated with renal function included hyponatremia (5%) and hypocalcemia (4%), while those associated with liver function included elevated lipase (8%) and alkaline phosphatase (5%).

The safety profile in this study was similar to that seen in prior GIST studies, with most AEs being mild or moderate in severity. Dr. Rutkoswki and his team of researchers note that “sunitinib appears to be generally well tolerated in patients with imatinib-resistant/ intolerant GIST who were ineligible for other sunitinib clinical trials.”

Gastrointestinal Stromal Tumor

Sunitinib in Phase III trial of Imatinib-resistant/ Intolerant GIST: Accounting for Crossover

Researchers have performed a novel statistical analysis to account for patient crossover in a double-blind, placebo-controlled phase III study of sunitinib (Sutent) in patients with imatinib-resistant or imatinib-intolerant gastrointestinal stromal tumor (GIST). Christopher Garrett, MD, M.D. Anderson Cancer Center, Houston, Texas, presented the results of the analysis, which confirmed the long-term overall survival benefit of sunitinib relative to placebo in patients with imatinib-resistant/intolerant GIST. The study has implications for other crossover trials. Sunitinib malate is an oral multitargeted tyrosine kinase inhibitor of various proteins, including KIT, PDGFRs and VEGFRs, and acts by inhibiting blood vessel growth and cell proliferation, starving off tumors.

At the 2006 annual ASCO meeting, researchers presented findings of a phase III multicenter clinical trial of sunitinib versus placebo showing that sunitinib led to four times longer progression-free survival (PFS) in GIST patients who had been previously treated with imatinib. Sunitinib also significantly improved overall survival (OS) (hazard ratio = 0.49). In the trial, 312 GIST patients were randomly assigned to 50 mg/day of sunitinib or placebo as a pill in six-week cycles with four weeks on and two weeks off treatment.

Based on results from a planned early analysis of the trial, the FDA granted accelerated approval to sunitinib in January 2006 for the treatment of patients with GIST whose disease progressed after imatinib treatment or who were initially imatinib-resistant. The preliminary results led to unblinding and 88% of patients in the placebo arm crossing over to the sunitinib group.

To account for crossover, Dr. Garrett and his colleagues used the rank-preserving structural failure time (RPSFT) method in addition to the log-rank test, Cox model, and Kaplan—Meier method. Ultimately, 243 patients were randomized to receive sunitinib while 118 patients received placebo. Of those patients in the placebo arm, 104 crossed over to sunitinib. By November 2007, the crossover had resulted in OS converging between the two treatment groups, with median OS for sunitinib versus placebo being 74.7 and 64.9 weeks, respectively. The RPSFT analysis, meanwhile, showed median OS of 36 weeks for placebo.

Common treatment-related grade 3/4 adverse events (AEs) in the sunitinib arm were fatigue (8%), hand—foot syndrome (4%), and hypertension (4%). Overall treatment-related AEs (all grades) were hypothyroidism (5%) and hypertension (14%). The incidence of cardiac AEs was 4%. Hematologic lab abnormalities included reduced hemoglobin (57%), neutrophils (55%), and platelets (41%), which were mostly grade 1/2 with similar frequency for short-term and extended therapy.