|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2002 American Society for Clinical Oncology Efficacy and Quality-of-Life Data Are Related in a Phase II Trial of Oral Chemotherapy in Previously Untreated Patients With Metastatic Colorectal CarcinomaByFrom the Mayo Clinic and Mayo Foundation, Rochester, and Duluth Community Clinical Oncology Program (CCOP), Duluth, MN; Illinois Oncology Research Association CCOP, Peoria, IL; Scottsdale CCOP, Scottsdale, AZ; Missouri Valley Cancer Consortium, Omaha, NE; Saskatoon Cancer Centre, Saskatoon, and Allan Blair Cancer Centre, Regina, Saskatchewan, Canada; Sioux Community Cancer Consortium, Sioux Falls, SD; and Toledo Community Hospital Oncology Program CCOP, Toledo, OH. Address reprint requests to Timothy J. Hobday, MD, Division of Oncology, Mayo Clinic, 200 First St SW, Rochester, MN 55905; email: hobday.timothy{at}mayo.edu
PURPOSE: To evaluate quality of life (QOL) and tumor response after administration of an oral chemotherapy regimen in patients with previously untreated metastatic colorectal cancer. PATIENTS AND METHODS: Seventy-eight patients received a mean number of 5.8 cycles of therapy. QOL data were analyzed at baseline, after every two cycles of therapy, and at the time of treatment discontinuation. The Uniscale and the European Organization for the Research and Treatment of Cancer Quality of Life Questionnaire C30 were both utilized. RESULTS: The confirmed response rate was 26% (95% confidence interval [CI], 17% to 37%). Median survival was 11.3 months (95% CI, 9.6 to 15.1 months). Global QOL scores were unchanged over the course of therapy by either tool. Only the physical function subscale score had worsened at the end of therapy. In an analysis of responding patients, significant and durable improvements in both global QOL measures as well as select subscale scores were observed. Diarrhea and physical function QOL scores had declined at the time of treatment discontinuation. Patients who did not respond to therapy had preserved QOL scores when they were evaluated after two cycles of therapy. CONCLUSION: This oral treatment strategy preserved QOL in treated patients. Global QOL measures as well as several QOL subscale scores significantly improved in patients with a documented response to therapy. The profile of improved QOL components indicated that patient well-being was related to tumor response in specific and perceivable ways. Nonresponding patients reported preserved QOL during the first two cycles of therapy. QOL analysis was feasible and informative in this moderately sized multicenter phase II trial.
METASTATIC COLORECTAL cancer is the second most common cause of cancer-related death in the United States.1 Fluorouracil (5-FU)-based chemotherapy has been a standard treatment of this disease for more than four decades, with response rates of 10% to 20% and a median survival of 11 to 12 months. The combination of 5-FU and irinotecan or oxaliplatin has improved outcomes for patients with metastatic colorectal cancer.2-4 However, intravenous, 5-FU–based chemotherapy remains an option for selected patients. Oral chemotherapy has practical and economic advantages over intravenous therapy. Patients prefer oral chemotherapy so long as benefit is not compromised.5 Oral 5-FU was initially abandoned for clinical application because of its poor and variable bioavailability.6 Oral eniluracil (ethynyluracil) is a mechanism-based complete inhibitor of dihydropyrimidine dehydrogenase, the rate-limiting catabolic enzyme for 5-FU metabolism. The use of attenuated doses of oral 5-FU with eniluracil has been investigated in phase I as well as phase II trials in metastatic colon cancer (including this study), with safety and efficacy documented.7-10 Two phase III trials comparing oral 5-FU/eniluracil with intravenous 5-FU and leucovorin as first-line treatment of metastatic colorectal cancer have been completed. Schilsky et al11 reported that the oral regimen did not meet statistical criteria for equivalence in terms of overall survival compared with intravenous 5-FU and leucovorin. Van Cutsem et al12 reported a statistically inferior overall survival for the oral regimen. Thus, further development of eniluracil has been terminated. Although designed to evaluate the antitumor efficacy of a 5-day regimen of oral 5-FU bracketed by a 7-day course of eniluracil in chemotherapy-naive patients with advanced colorectal cancer, this trial was one of the initial North Central Cancer Treatment Group (NCCTG) phase II studies to incorporate quality-of-life (QOL) assessment. Recognizing the need to determine the feasibility and utility of evaluating QOL in multicenter clinical trials, we added a prospective QOL component to this moderately sized phase II trial. The evaluation of QOL in phase II trials in advanced cancer potentially adds important information to the evaluation of the proposed therapy and its suitability for use in the palliative setting. QOL analysis in phase II trials may also facilitate QOL evaluation in subsequent phase III trials, providing insight regarding potential logistical issues as well as generating hypotheses. We present a summary of response, toxicity, and patient outcome in our phase II trial with an emphasis on the feasibility and usefulness of the QOL analysis.
Patient Eligibility All patients had histologic or cytologic evidence of recurrent or metastatic colorectal adenocarcinoma that was not resectable with curative intent. Eligible patients 18 years or older had to have measurable disease, clear evidence of progression if the only disease site had been previously irradiated, and an Eastern Cooperative Oncology Group performance status of 0, 1, or 2. Life expectancy had to be more than 12 weeks. Laboratory eligibility criteria included adequate hematologic reserve, renal function, and hepatic function. All patients gave informed consent to participate in the study, which was approved by local institutional review boards.
Patients were excluded if they had received prior 5-FU in any setting, prior radiation therapy to
Patient Evaluation At study entry, before every other course of treatment, and at discontinuation of 5-FU/eniluracil therapy, patients were to complete two QOL instruments. Patients completed the QOL instruments before consulting with their treating physician regarding their symptomology and disease status. One of the QOL instruments utilized was the Uniscale (UNI), which is a visual analog, single-item questionnaire intended to measure global QOL.13 The other QOL instrument was the European Organization for Research and Treatment of Cancer (EORTC) Quality of Life Questionnaire (QLQ) C30, version 2.0, a 30-item questionnaire that includes a two-item global status QOL subscale, five single-item and multi-item functional subscales (physical, role, emotional, social, and cognitive functioning), and nine single-item and multi-item symptom subscales (fatigue, nausea and vomiting, pain, dyspnea, insomnia, anorexia, constipation, diarrhea, and financial difficulties).14,15 This tool was included so that a detailed profile of QOL data could be obtained to supplement the UNI single-item global score.
Treatment Plan All toxicity evaluations were graded according to the NCCTG/National Cancer Institute common toxicity criteria, version 1, unless otherwise specified. Any grade 2 to 4 toxicity during the 7 days of drug administration resulted in discontinuation of remaining doses in that cycle. Appropriate dose modifications were mandated for significant toxicity.
Response Criteria Patients who died (or who were lost to follow-up) without progression were considered to have progressed at the date of death (or last contact) unless documentation proved otherwise, in which case they were considered to have no progression at the date of last tumor evaluation. Duration of response was calculated from the date of the patient’s first objective status of CR or PR to the date of progression. Time to progression was calculated from study entry to disease progression. Overall survival was calculated from the date of study entry to death or last contact.
Statistical Considerations For reference, we note that this study was designed to detect a confirmed tumor response rate of at least 25% versus a null hypothesis of inactivity (ie, response rate < 10%) using a two-stage, phase II Simon20 design. Confirmed tumor response was defined as having an objective tumor response of PR or CR on at least two consecutive tumor evaluations. Three confirmed responses in the initial 21 assessable patients warranted expanding accrual to an additional 45 patients. Eleven confirmed responses in all 66 assessable patients was sufficient evidence of efficacy. Other secondary end points included toxicity, duration of response, time to disease progression, and survival. Nonhematologic toxicity is reported as the maximum grade for each individual toxicity. Kaplan-Meier methodology21 was used to describe the distribution of duration of response, time to disease progression, and survival. Cox proportional hazards modeling22 was used to identify factors (eg, age, sex, PS) significantly associated with duration of response, time to disease progression, and survival. Graphical methods confirmed underlying statistical assumptions regarding proportional hazards. P values less than .05 were considered statistically significant. Adjustments to statistical tests were not made for multiple comparisons.
Patient Characteristics A total of 79 patients were registered onto this study between May 19, 1997, and June 24, 1998, from 18 NCCTG member institutions. One patient was deemed ineligible because there was no histologic proof of metastatic adenocarcinoma despite clinical notation of liver metastases at surgery. Table 1 lists the characteristics for the remaining 78 eligible patients. At study entry, eligible patients ranged in age from 34 to 80 years (mean age, 62.4 years), 46% had a PS of 0, 96% were white, and 41% were female. Twenty-nine percent of eligible patients presented with histologic grade 1 or 2 disease. Six percent of patients had received prior radiation therapy.
Patient Outcome All 78 eligible patients are considered assessable for response (Table 2). A total of 20 confirmed responses were observed (1CR, 19 PRs). The estimated confirmed response rate was 26%. All 20 patients who achieved a CR or PR subsequently progressed, with an estimated median duration of response of 6.6 months. Overall, 96% of patients have progressed and 77% have died, with a minimum follow-up of 12.4 months for surviving patients. Median survival was 11.3 months (Fig 1). The estimated 6-month and 1-year progression-free rates were 39% and 9%, respectively; the estimated 6-month and 1-year survival rates were 80% and 49%, respectively. Patient characteristics (ie, age, sex, PS, tumor grade, and prior radiation therapy) were not significantly associated with objective response, duration of response, time to disease progression, or survival.
Toxicity On average, patients completed 5.8 courses of treatment (range, one to 28 courses). A majority (91%) discontinued therapy because of disease progression; the remainder went off therapy because of toxicity or refusal. Severe toxicity for course 1 and over all courses of treatment is listed in Table 3. Grade 3 or 4 myelosuppression consisted primarily of neutropenia, which was observed in 50% of patients, with 24% experiencing grade 4 toxicity. Nonhematologic toxicity included primarily diarrhea, nausea, vomiting, lethargy, infection, and neurotoxicity. Grade 3 or 4 diarrhea occurred in 19% of patients. Five episodes of grade 3 cerebellar toxicity were observed.
QOL Compliance and Completeness Among all patients, 248 (88%) of a possible 283 EORTC QLQ-C30 forms were completed. The primary reason for an uncompleted form was personnel neglecting to give the patient the form (11% of cases [31 of 283]). Other reasons for noncompletion of the 30-item form included the patient having died, being too sick, or having had no contact. Three of four patients requiring assistance to complete the form needed help at their first experience with the questionnaire (ie, baseline) because of a recent surgery, the patient being weak, and the spouse noting the responses for the patient. Initially, patients completed the 30-item questionnaire in an average time of 11.1 minutes (range, 3 to 40 minutes). Patients requiring more than 20 minutes to complete the longer form did so at baseline (ie, first experience). On average, patients completed one less set of QOL tools versus the expected number to be administered (range, four less to one extra). Forty-one percent (32 of 78) of patients completed exactly the number of sets of QOL forms required. Seven patients completed more of one tool than the other; more often the UNI was not completed. Missing data were sparse and occurred at each reporting period, averaging less than two forms missing at a given reporting period. More importantly, reports from the treating sites indicated no perceivable burden noted on either patients or the clinical environment. One or both QOL instruments were completed at baseline by 77 of the 78 patients. Eleven patients did not complete a subsequent QOL assessment and were excluded from the analysis presented. Imputation methods using both the last value carried forward and zero values carried forward were used for cross-validation, and the results were consistent with the analysis presented.19 Rapid disease progression (eight patients) or death (one patient) during the first or second treatment cycle accounted for the majority of patients not included in the QOL analysis.
Correlation of QOL With Clinical End Points
In exploratory analysis, we further classified patients as either having achieved a tumor response (CR or PR) or nonresponders (Table 4 and Fig 3). At the first reporting period, clinically significant improvements in mean QOL scores were seen for both global tools (13-point increase in UNI, 22-point increase in global QOL subscale of EORTC QLQ-C30) as well as the subscales of social functioning, emotional functioning, nausea and vomiting, anorexia, constipation, insomnia, and pain among responding patients. These improvements carried through to the end of the study treatment for the global QOL, emotional functioning, anorexia, and insomnia subscales (Table 5). The other subscale scores returned toward baseline by the time progression was evident. Diarrhea was the only subscale in which scores were observed to worsen during the first two cycles of treatment. Physical functioning declined significantly from the initiation of therapy to the time of discontinuation. Equally as important was the finding that subscale scores for cognitive function, fatigue, and dyspnea did not show any overall change.
We illustrate the changes in global QOL for each patient using a method known as an event chart.19,23 This chart graphs changes in QOL score for each patient over time, identifying clinically significant changes by different line-hatching patterns. Figure 4 illustrates that the mean improvement in QOL scores for responding patients was not due to large improvements in just a few patients but to improvements in QOL for virtually all of the responding patients. Differences between responders and nonresponders are evident by the markedly different appearance of the event chart lines between the two groups. We constructed a similar chart illustrating declines in QOL instead of improvements that mirrored results depicted in Fig 4 (data not shown).
QOL scores did not change significantly from baseline to the first reporting period for patients classified as nonresponders, with the exception of worsening fatigue. QOL as measured by the UNI did decline significantly from baseline to the time of treatment discontinuation, and the global subscale of the EORTC QLQ-C30 decreased an average of 9 points. A decline in QOL scores was also seen in the subscale scores for role functioning, social functioning, and physical functioning at the time of treatment discontinuation.
This multicenter phase II trial of a 5-day regimen of oral 5-FU with eniluracil administered to patients with untreated advanced colorectal cancer adds to a body of literature regarding the safety and efficacy of this combination. Therapy was generally well-tolerated, with neutropenia, diarrhea, and neurologic abnormalities as the most prominent adverse affects. QOL analysis suggests that 5-FU/eniluracil chemotherapy did not adversely affect global QOL or any particular QOL subscale. Marked differences in a subset of specific QOL end points were seen between those patients whose tumors responded to treatment and those whose tumors did not. Some of these differences carried through to treatment discontinuation. These data indicate that QOL data may be able to serve as surrogates for tumor response. We recognize that in the phase II setting, it is difficult to arrive at definitive conclusions regarding the effect of chemotherapy on QOL. The lack of a randomized comparison arm makes it difficult to determine whether declining QOL scores over time are due to the treatment, progressive metastatic disease, or both. In our study, 91% of patients discontinued therapy because of disease progression. This factor may largely be responsible for any declining QOL scores from pretreatment to the time of treatment discontinuation. However, there are several reasons to infer that QOL was not significantly adversely affected by 5-FU/eniluracil. Only 9% of patients discontinued treatment because of toxicity or refusal. There was no clinically significant decline in QOL scores for all patients over the length of the study, with the exception of physical functioning. On the contrary, for patients who responded to the chemotherapy, there was significant improvement of as much as three SDs in global QOL scores over the first two cycles of therapy by both measures, as well as in several subscales (Table 3). At the time of treatment discontinuation, global QOL, emotional functioning, and symptom improvements in anorexia and insomnia remained significantly improved over baseline, which suggests that further cycles of therapy remained well tolerated. The apparent decline in QOL from pretreatment to treatment discontinuation in nonresponding patients is likely due to a combination of disease-related symptoms and treatment side effects. However, even in these patients, only fatigue significantly worsened over the first two cycles of therapy. Our conclusion is that the QOL analysis, supported by the toxicity results, suggests that treatment with 5-FU/eniluracil is well tolerated and does not seem to significantly adversely affect QOL, even among patients who do not respond to the treatment. Responders can expect to experience a marked improvement in a fairly specific subset of QOL end points. Further, improvement in this specific subset of QOL end points may be indicative of the initiation of tumor response. Further research is needed to ascertain whether the QOL changes are concurrent with or predictive of tumor response. The data from this trial support the feasibility of collecting and assessing QOL in multicenter phase II trials. We observed a high compliance rate (88%), despite the limited sample size (n = 78), high number of membership institutions involved (n = 18), and one of the instruments utilized having 30 items. Furthermore, identifying important QOL end points in a phase II study in this fashion may serve to facilitate QOL analysis in future studies of the agent(s). The important subsets of QOL end points should be measured in larger phase III trials to obtain more precise estimates of treatment effects and adequate statistical power to detect such effects of treatment on QOL. The fact that some of the observed differences in this study did not reach statistical significance may be due, at least in part, to sample size limitations. The event chart is a relatively new method used in this study to provide insight into the pattern of changes in self-reported QOL over time per patient. The major conclusion from these data (reported in Fig 4) is that mean improvements over time in the responders group were not due to large improvements in only a few patients but rather reflect QOL improvements in virtually all of these patients. It is also interesting to note the pattern of QOL changes for some of the responding patients. Specifically, for 11 of the 20 responders, global QOL (as assessed by the QLQ-C30) was significantly improved over baseline values and remained so until the patient went off study. This is, in a sense, surprising, in that one would expect a decline in self-reported QOL just before or around the time of disease progression or cessation of treatment because of disease-related symptoms. The inclusion of QOL questionnaires in phase II studies has been a subject of some controversy in the literature.16,17,23 Some have suggested that QOL analysis adds an extra burden to phase II trials and is of limited value because of the lack of a control group. In our experience, the burden to the patients and the demand on system resources are not substantial. Using a minimalist approach, we have been able to successfully obtain QOL data in numerous NCCTG phase II trials.17,19 The analysis and interpretation of such data are no more complicated than any other secondary end point, such as results of laboratory assays or genetic testing,24 and are particularly relevant in a setting where treatment is palliative. Moreover, we believe phase II trials are a cost-effective platform for assessing the relevance and logistics for QOL data in developing new anticancer therapies. This allows for a refined and efficient QOL assessment to be carried out in subsequent phase III trials. Waiting until the initiation of a phase III trial likely involving hundreds of patients to validate a strategy by which to address questions about patient QOL is problematic in our opinion. Indeed, we are investigating routine incorporation of QOL assessment in selected phase I trials. In this phase II trial, analysis of QOL scores suggested that after two cycles of therapy, QOL was maintained, and at the conclusion of therapy, QOL was not clinically significantly affected by the protocol treatment. When QOL results for patients who responded to therapy were analyzed, there were clear improvements in QOL over the first two cycles of therapy. We believe that this type of analysis provides important insight into the tolerability of the therapy. Any subsequent decline or return to baseline of QOL scores becomes more difficult to interpret, as the relative contributions of chronic treatment-related toxicity and progressive metastatic disease are difficult to separate. One of our original goals for this study was to determine whether QOL analysis in phase II studies in advanced cancers yields worthwhile information. We believe the analysis of this trial did provide important information regarding the effect of oral 5-FU and eniluracil on patient QOL, and we plan to include QOL analysis in future phase II trials of advanced cancers.
APPENDIX
This study was conducted as a collaborative trial of the North Central Cancer Treatment Group and Mayo Clinic and was supported by United States Public Health Service grant nos. CA-25224, CA-37404, CA-15083, CA-63826, CA-60276, CA-35113, CA-35269, CA-63849, CA-35103, CA-35415, CA-35195, CA-52352, CA-37417, CA-35103, CA-35101, CA-35448, and CA-35272.
1. Boring CC, Squires TS, Tong T: Cancer statistics 1993. CA Cancer J Clin 43: 7-26, 1993[Medline] 2. Douillard JY, Cunningham D, Roth AD, et al: Irinotecan combined with fluorouracil compared with fluorouracil alone as first-line treatment for metastatic colorectal cancer: A multicentre randomised trial. Lancet 355: 1041-1047, 2000[CrossRef][Medline]
3. Saltz LB, Cox JV, Blanke C, et al: Irinotecan plus fluorouracil and leucovorin for metastatic colorectal cancer. N Engl J Med 343: 905-914, 2000
4. de Gramojnt A, Figer A, Seymour M, et al: Leucovorin and fluorouracil with or without oxaliplatin as first-line treatment in advanced colorectal cancer. J Clin Oncol 18: 2938-2947, 2000
5. Liu G, Franssen E, Fitch MI, et al: Patient preferences for oral versus intravenous palliative chemotherapy. J Clin Oncol 15: 110-115, 1997 6. Hahn RG, Moertel CG, Schutt AJ, et al: A double-blind comparison of intensive course 5-fluorouracil by oral vs. intravenous route in the treatment of colorectal cancer. Cancer 35: 1031-1035, 1975[CrossRef][Medline] 7. Baker SD, Khor SP, Adjei AA, et al: Pharmacokinetic, oral bioavailability, and safety study of fluorouracil in patients treated with 776C85, an inactivator of dihydropyrimidine dehydrogenase. J Clin Oncol 14: 3085-3096, 1996[Abstract]
8. Schilsky RL, Hohneker J, Ratain MJ, et al: Phase I clinical and pharmacologic study of eniluracil plus fluorouracil in patients with advanced cancer. J Clin Oncol 16: 1450-1457, 1998
9. Schilsky RL, Bukowski R, Burris H 3rd, et al: A multi-center phase II study of a five-day regimen of oral 5-fluorouracil plus eniluracil with or without leucovorin in patients with metastatic colorectal cancer. Ann Oncol 11: 415-420, 2000
10. Mani S, Hochster H, Beck T, et al: Multi-center phase II study to evaluate a 28-day regimen of oral fluorouracil plus eniluracil in the treatment of patients with previously untreated metastatic colorectal cancer. J Clin Oncol 18: 2894-2901, 2000
11. Schilsky RL, Levin J, West WH, et al: Randomized, open-label, phase III study of a 28-day oral regimen of eniluracil plus fluorouracil versus intravenous fluorouracil plus leucovorin as first-line therapy in patients with metastatic/advanced colorectal cancer. J Clin Oncol 20: 1519-1526, 2002 12. Van Cutsem E, Sorensen J, Cassidy J, et al: International phase III study of oral eniluracil (EU) plus 5-fluorouracil (5-FU) versus intravenous (IV) 5-FU plus leucovorin in the treatment of advanced colorectal cancer (ACC). Proc Am Soc Clin Oncol 20: 131a, 2001 (abstr 522) 13. Spitzer WO, Dobson AJ, Hall J, et al: Measuring the quality of life of cancer patients: A concise QL-index for use by physicians. J Chronic Dis 34: 585-597, 1981[CrossRef][Medline] 14. Selby PJ, Chapman JA, Etazadi-Amoli J, et al: The development of a method for assessing the quality of life of cancer patients. Br J Cancer 50: 13-22, 1984[Medline]
15. Aaronson NK, Ahmedzai S, Bergman B, et al: The European Organization for Research and Treatment of Cancer QLQ-C30: A quality-of-life instrument for use in international clinical trials in oncology. J Natl Cancer Inst 85: 365-376, 1993 16. Sloan JA, Loprinzi CL, Kuross SA, et al: Randomized comparison of four tools measuring overall quality of life in patients with advanced cancer. J Clin Oncol 16: 3662-3673, 1998[Abstract] 17. Sloan JA, O’Fallon JR, Suman VJ, et al: Incorporating quality of life measurement in oncology clinical trials. Proc Am Stat Assoc 282-287, 1998 18. Duffy DE, Santner TJ: Confidence intervals for a binomial parameter based on multistage tests. Biometrics 43: 81-93, 1987[CrossRef] 19. Atherton PJ, Novotny PJ, Jasperson B, et al: So what happened to all the patients? Event charts for summarizing clinically significant changes in oncology quality of life data. Proc Am Soc Clin Oncol 20: 398a, 2001 (abstr 1590) 20. Simon R: Optimal two-stage designs for phase II clinical trials. Control Clin Trials 10: 1-10, 1989[Medline] 21. Kaplan E, Meier P: Nonparametric estimation from incomplete observation. J Am Stat Assoc 53: 457-481, 1958[CrossRef] 22. Cox DR: Regression models and life tables (with discussion). J R Stat Soc 74: 187-220, 1972 23. Sloan JA, Novotny PJ, Loprinzi CL: Design and analysis of cancer control studies using the SAS system. Proc SUGI 25: 259-325, 2000 24. Sloan JA, Varricchio C: Quality of life endpoints in prostate chemoprevention trials. Urology 57: 235-240, 2001 (suppl 1)[CrossRef][Medline] Submitted August 3, 2001; accepted August 5, 2002.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2002 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
25% of their bone marrow, prior radiotherapy in the previous 4 weeks, prior thoracic or abdominal surgery in the previous 3 weeks, or prior flucytosine in the previous 2 weeks. Patients had to be able to swallow and retain oral medication and were to be free from uncontrolled infection or other severe medical condition that would, in the judgment of the investigator, make them inappropriate for the study. Pregnant or lactating women, men or women with childbearing potential not using effective contraception, patients with CNS metastasis, and patients with another active malignancy were also excluded. Patients with evidence of malabsorption or with a history of major resection of the stomach, proximal small bowel, or the terminal ileum affecting gastrointestinal function were excluded. 
