|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 22, No 8 (April 15), 2004: pp. 1439-1446 © 2004 American Society of Clinical Oncology. DOI: 10.1200/JCO.2004.10.043 Relationship of Baseline Serum Bilirubin to Efficacy and Toxicity of Single-Agent Irinotecan in Patients With Metastatic Colorectal CancerFrom the Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA; Pfizer Oncology, New York, NY; and Committee on Clinical Pharmacology and Pharmacogenomics, Department of Medicine and Cancer Research Center, University of Chicago, Chicago, IL Address reprint requests to Jeffrey A. Meyerhardt, MD, Dana-Farber Cancer Institute, 44 Binney St, Boston, MA 02115; e-mail: jmeyerhardt{at}partners.org
PURPOSE: To examine the predictive value of baseline serum bilirubin measurement for chemotherapy-related toxicity or efficacy among patients receiving irinotecan for metastatic colorectal cancer. METHODS: We performed a secondary analysis of a cohort of 287 patients treated in a multicenter, phase III study with single-agent irinotecan administered either weekly or once every 3 weeks. Patients were grouped into three categories of baseline bilirubin measurements (0 to 0.4, 0.5 to 0.9, and 1.0 to 1.5 mg/dL). We performed analyses of overall survival, time to progression, and treatment-related toxicity based on bilirubin category, as well as using bilirubin as a continuous variable. RESULTS: With a median follow-up of 15.8 months, baseline serum bilirubin was not predictive of 1-year survival (42.4%, bilirubin 0 to 0.4; 42.3%, bilirubin 0.5 to 0.9; 48.1%, bilirubin 1.0 to 1.5 mg/dL), median overall survival (10.1, 9.7, and 15.6 months, respectively; P = .5), or median time to progression (2.8, 3.0, and 4.1 months, respectively; P = .5). Patients with elevated bilirubin had a significantly greater risk grade 3 to 4 neutropenia; however, this was limited to patients treated on a weekly schedule (P trend = .03) and not once every 3 weeks (P trend = .8). Other toxicities were not significantly different by initial bilirubin measurement. CONCLUSION: Although modest elevations of bilirubin (1.0 to 1.5 mg/dL) are associated with increased grade 3 to 4 neutropenia in patients treated with weekly irinotecan, baseline serum bilirubin does not reliably predict overall irinotecan-related toxicity or efficacy. Additional methods, including potential application of pharmacogenetic information, are needed to optimize irinotecan dosing and tailor therapy to individual patients.
Irinotecan (CPT-11) is a camptothecin-derived prodrug inhibitor of the enzyme topoisomerase I that is active against a variety of malignancies, including colorectal, lung, gastric, esophageal, ovarian, and head and neck carcinomas.1 The metabolism and disposition of irinotecan are complex, with hydrolysis of the parent prodrug by hepatic carboxylesterases to the active metabolite (SN-38), detoxification of SN-38 to an inactive glucuronidated form (SN-38G) by uridine diphosphate glucuronosyltransferase isoform 1A1 (UGT1A1), and excretion of SN-38G in urine and bile.2 In addition, several other metabolites of irinotecan are formed by oxidative metabolism of irinotecan by the P450 enzymes, CYP3A4 and CYP3A5.3 Although it is an efficacious agent, irinotecan can cause significant diarrhea and bone marrow suppression in certain patients. However, the severity of the drug’s toxicities is largely unpredictable. The processes that mediate drug activation and elimination, including interactions with other medications, may have a substantial influence on this interpatient variability.4 Mathijssen et al3 found that body-surface area as well as other tested body size measures were unrelated to irinotecan metabolism. Studies are now underway to investigate individualization of irinotecan dosing based on genotypic analysis of the various drug metabolizing enzymes and drug transport proteins. For example, recent studies have demonstrated that genetic variability at the UGT1A1 promoter correlates with both pharmacokinetics and toxicity.5,6 Both SN-38 and bilirubin undergo glucuronidation by UGT1A1. A significant correlation was observed between SN-38 glucuronidation rates and bilirubin glucuronidation rates by human liver microsomes.7 In a case report, two patients with Gilbert’s syndrome, a deficiency of UGT1A1, experienced more severe toxicities from irinotecan.8 Two prior studies suggested that baseline serum bilirubin was predictive of neutropenia but not diarrhea or efficacy.9,10 However, conclusions that can be drawn from all of these studies are limited because of the nonuniform nature of patient selection, variable irinotecan dosing schedules, and lack of detailed information regarding tolerated levels of bilirubin. We therefore examined the influence of baseline serum bilirubin on irinotecan-related toxicity and efficacy in 291 patients participating in a multicenter randomized study of two dose schedules of single-agent irinotecan (weekly or once every 3 weeks) for fluorouracil (FU)-refractory colorectal cancer.11 By studying patients who were enrolled onto a prospective clinical trial, we were able to control for other clinical predictors of outcome and eliminate the influence of other concomitantly delivered chemotherapeutic agents.
Study Design and Patient Selection This secondary analysis used a cohort of 291 patients treated during a multicenter, open-label, randomized study comparing two dosage regimens of irinotecan (Camptosar; Pfizer Oncology, New York, NY) in patients with metastatic colorectal cancer who experienced disease progression or recurrence after receiving FU-based chemotherapy. The study design and results have been published recently.11 Briefly, patients were required to have measurable, biopsy-proven colorectal cancer and to have experienced either disease progression after receiving FU-based chemotherapy or relapse within 12 months of adjuvant therapy. Several eligibility criteria of the original trial that are particularly important to note for this secondary analysis were that inclusion criteria included an initial serum bilirubin  1.5 mg/dL, AST 3 x upper level of institutional normal ( 5 x upper level of institutional normal if liver metastases were present), and no known history of Gilbert’s syndrome.
Patients enrolled onto the trial were prospectively stratified by age (< 70 v
Bilirubin Measurements
Statistical Analysis As a requirement of the treatment trial, safety assessments and CBCs were performed weekly. All adverse events were reported according to the National Cancer Institute Common Toxicity Criteria.14 We examined toxicities by baseline bilirubin measurements using Mantel-Haenszel for trend. Given the differences in toxicity profiles between the regimens of administration weekly and once every 3 weeks, primary analyses were stratified by treatment regimen. Logistic regression modeling was used to adjust for potential confounders of significant toxicities, adjusted for age, performance status, prior pelvic irradiation, and sex. Bilirubin was similarly entered into these models with indicator variables and as a continuous variable. All analyses were performed with SAS version 8.2 (SAS Institute, Cary, NC).
Baseline Characteristics by Baseline Bilirubin Between August 1998 and January 2001, 291 patients with metastatic colorectal cancer were treated with single-agent irinotecan, either weekly or every 3 weeks. From this cohort, 287 patients had baseline bilirubin measurements prospectively recorded between 0.0 and 1.5 mg/dL. Table 1 shows the distribution of other baseline patient characteristics according to predefined categories of baseline serum bilirubin. Male patients had a significantly greater initial bilirubin level, as previously reported.15,16 However, bilirubin was not significantly related to age, race or ethnicity, baseline performance status, primary site of tumor, baseline carcinoembryonic antigen level and hemoglobin, or treatment arm. In addition, there were no appreciable differences by prior pelvic irradiation, prior adjuvant FU, prior FU for metastatic disease, number of organs involved, or baseline WBC count (data not shown).
Efficacy of Irinotecan by Baseline Bilirubin We tested whether modest elevations in bilirubin (up to 1.5 mg/dL) influenced the efficacy of single-agent irinotecan in the treatment of metastatic colorectal cancer. Figure 1 shows the Kaplan-Meier distributions for overall survival by bilirubin category for the entire cohort. There were no statistically significant differences in 1-year survival (OS) or median OS (Table 2) on the basis of baseline bilirubin measurement. Similarly, there were no differences in TTP by the three categories of bilirubin (Fig 2 and Table 2). When we excluded patients in the arm receiving treatment every 3 weeks whose initially dose of irinotecan was 300 mg/m2, these results did not change (P = .4 and P = .5 for OS and TTP, respectively).
Using Cox proportional hazards models, we adjusted for potential confounders of the relationship between baseline serum bilirubin and efficacy. For overall mortality, we entered serum bilirubin into the model according to our predefined categories (using 0 to 0.4 mg/dL as the referent group) and, separately, as a continuous variable. After adjustment for age, baseline performance status, hemoglobin and WBC count, number of organs involved, and treatment arm, there were no significant differences in mortality for patients with an initial bilirubin of 0.5 to 0.9 mg/dL (hazard ratio [HR], 1.01; 95% CI, 0.74 to 1.38) or 1.0 to 1.5 mg/dL (HR, 0.95; 95% CI, 0.58 to 1.54) when compared with patients in the lowest category (P trend = .4). With the exception of age, stratified analyses did not demonstrate subgroups with consistent trends toward differences in overall mortality by baseline bilirubin (Table 3). Among patients at least 70 years old, increased bilirubin seemed to be predictive of worse overall mortality, although the frequency of bilirubin  1 mg/dL was considerably reduced compared with patients younger than 70 years old (six of 96 [6%] for patients 70 compared with 27 of 191 [14%] for patients < 70 years old). When the cross-product of age (dichotomized) and bilirubin (by categories) was entered into the proportional hazards model, a significant interaction was confirmed (P = .01). Nonetheless, bilirubin was not associated with TTP in these age subgroups. Similarly, there was no statistical interaction between bilirubin and treatment arm (P = .3), sex (P = .6), race or ethnicity (P = .7), or baseline performance status (P = .3).
Similarly, after we adjusted for other potential confounders, TTP was not significantly different for patients with a bilirubin level of 0.5 to 0.9 mg/dL (HR, 0.98; 95% CI, 0.72 to 1.35) or 1.0 to 1.5 mg/dL (HR, 0.70; 95% CI, 0.43 to 1.14) when compared with those in the referent group. Male patients and African-American patients did appear to have improved TTP with increased baseline bilirubin measurements, although caution should be exercised in interpreting these results given the small sample sizes and multiple comparisons. When we entered the cross-product between bilirubin (by categories) and each of the strata (dichotomized) listed in Table 3 in the proportional hazards model, the P values were not significant (age, P = .5; treatment arm, P = .4; sex, P = .9; race, P = .8; and baseline performance status, P = .1). These findings for overall mortality or TTP did not appreciably change when patients in the arm initially treated with 300 mg/m2 of irinotecan once every 3 weeks were excluded (data not shown).
Toxicity of Irinotecan by Baseline Bilirubin
We further analyzed irinotecan-related toxicities among the entire cohort regardless of treatment arm. Again, only grade 3 or 4 neutropenia correlated with baseline bilirubin level (29%, bilirubin 0 to 0.4; 31%, bilirubin 0.5 to 0.9; and 46%, bilirubin 1.0 to 1.5 mg/dL; P trend = .002 after adjusting for age, performance status, sex, prior pelvic irradiation, and treatment arm). To further explore the relationship between baseline bilirubin measurement and irinotecan-induced myelosuppression, we plotted the log of nadir absolute neutrophil counts (ANCs) against pretreatment total bilirubin level (Fig 3). For these analyses, the lowest ANC from any cycle for each given patient was used. As demonstrated, the correlation between bilirubin and log of the nadir ANC was fairly low (–0.19 for all patients, –0.29 for patients treated weekly, and –0.17 for patients treated once every 3 weeks). When we restricted the data to nadir ANC during cycle 1, the correlation with bilirubin was even weaker (–0.14 [P = .03] for all patients, –0.12 [P = .26] for patients treated weekly, and –0.14 [P = .06] for patients treated once every 3 weeks).
Analysis of Dose Reductions We analyzed proportions of patients requiring dose reductions during the first 4 weeks of therapy based on treatment arm and baseline bilirubin measurement (Fig 4). All dose reductions were based on standardized, prespecified parameters within the protocol. For arms of patients treated weekly and once every 3 weeks, serum bilirubin was associated with a nonsignificant elevation in the likelihood of dose reductions. At week 4 of the first treatment cycle in the arm with patients treated weekly, full-dose irinotecan was received by 31% of patients with a bilirubin level of 1.0 to 1.5 mg/dL, compared with 46% of patients with a bilirubin 0 to 0.4 mg/dL and 30% of patients with a bilirubin 0.5 to 0.9 mg/dL (P = .37). Similarly, within the arm with patients treated once every 3 weeks, by the fourth week, 50% of patients required a dose reduction with an initial bilirubin 1.0 to 1.5 mg/dL, compared with 37% of patients with bilirubin less than 0.5 mg/dL and 36% of patients with bilirubin 0.5 to 0.9 mg/dL (P = .14).
Using a cohort of patients with metastatic colorectal cancer treated in a multicenter, randomized trial of irinotecan, we were able to study the influence of baseline serum bilirubin measurements on both efficacy and toxicity. Modest elevations (up to 1.5 mg/dL) of bilirubin did not influence the efficacy of irinotecan in this patient population. Bilirubin was significantly predictive of grade 3 to 4 neutropenia; however, this relationship was principally seen among patients treated on the weekly schedule. In contrast, elevations of bilirubin within the normal range did not consistently influence other irinotecan-associated toxicities. Independent of irinotecan dosing schedule, baseline bilirubin was not predictive of treatment efficacy in our cohort. An increasing number of malignancies have been found to be responsive to irinotecan therapy. However, the drug has an unpredictable toxicity profile, with occurrences of severe neutropenia and diarrhea causing particular concern. The complex metabolism of irinotecan, with various enzymes and proteins involved in the activation and excretion of the drug, likely contributes to its variability. Although pharmacogenetics offers promise to be able to better tailor therapies to the individual patient, other potential predictors of outcomes and toxicities are of great interest. Because UGT1A1 is responsible for both detoxification of SN-38 and bilirubin glucuronidation, we were interested in studying whether bilirubin levels would serve as a useful marker in treating patients with irinotecan.
Prior studies have found that patients with hepatic dysfunction and serum bilirubin levels greater than 3 x the upper limit of normal have exponential decreases in the clearance of irinotecan and increased dose-limiting neutropenia.17,18 However, the majority of patients may have only modest increases in serum bilirubin. Innocenti et al19 recently presented data to suggest a correlation between pretreatment bilirubin level and UGT1A1 polymorphisms. In addition, in their cohort of 63 patients, the seven patients with bilirubin more than 0.8 mg/dL had significantly lower nadir ANC levels than patients with bilirubin An advantage of the cohort of patients used in our study is that patients were treated on two dose schedules of irinotecan.11 In the treatment of patients with colorectal cancer, irinotecan is given either as a single agent or in combination regimens either weekly,20,21 every 2 weeks,22 or every 3 weeks.23 The toxicity profile of each schedule differs and, thus, predictive factors for irinotecan-related side effects should be studied in the context of frequency of therapy. Given the influence of serum bilirubin on grade 3 to 4 neutropenia among patients treated on a weekly schedule, patients with modest bilirubin elevations may be better served by treatment with a less frequent dosing schedule. Moreover, patients with elevations in bilirubin appeared to require fewer dose reductions on the schedule of treatment once every 3 weeks than on the weekly regimen; however, this did not reach statistical significance. This study has several limitations. First, we only have data on pretreatment bilirubin levels and do not have data on germline polymorphisms in the various enzymes responsible for irinotecan metabolism and excretion. Although pretreatment bilirubin may be correlated with UGT1A1 functionality,19 bilirubin elevations may also be reflective of invasive disease in the liver as well as other factors. Second, given that patients were all treated with single-agent irinotecan, we would suggest caution when applying these data to patients treated with combination regimens. Finally, because of protocol requirements, all patients had baseline bilirubin measurements of no greater than 1.5 mg/dL. Thus, the impact of higher bilirubin levels on the efficacy and toxicity of irinotecan could not be examined in this study. In conclusion, pretreatment bilirubin levels within the normal range have limited value in clinical decision making when using irinotecan. These data suggest that clinicians who prefer a weekly schedule of irinotecan need to be cautious of significant neutropenia in patients with elevated bilirubin. However, more effective predictors of efficacy and toxicity are still needed. It is hoped that greater understanding of the pharmacogenetics of irinotecan metabolism will allow for more rational choices in treatment and dosing schedules.
The following authors or their immediate family members have indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. Owns stock (not including shares held through a public mutual fund): Mark Ratain, OGTIAI Polymorphism, Noveco; James McGovren, Pfizer. Acted as a consultant within the last 2 years: Mark Ratain, Noveco. Received more than $2,000 a year from a company for either of the last 2 years: Ambrose Kwok, Pfizer; James McGovren, Pfizer; Charles Fuchs, Pfizer.
Supported by Pfizer Oncology, New York, NY. Dr Meyerhardt is supported in part by a K07 award from the National Cancer Institute (1K07CA097992-01A1) and an American Society of Clinical Oncology career development award. Authors’ disclosures of potential conflicts of interest are found at the end of this article.
1. Pizzolato JF, Saltz LB: The camptothecins. Lancet 361:2235-2242, 2003[CrossRef][Medline] 2. Klein CE, Gupta E, Reid JM, et al: Population pharmacokinetic model for irinotecan and two of its metabolites, SN-38 and SN-38 glucuronide. Clin Pharmacol Ther 72:638-647, 2002[CrossRef][Medline]
3. Mathijssen RH, van Alphen RJ, Verweij J, et al: Clinical pharmacokinetics and metabolism of irinotecan (CPT-11). Clin Cancer Res 7:2182-2194, 2001
4. Ratain MJ: Irinotecan dosing: Does the CPT in CPT-11 stand for "Can’t Predict Toxicity"? J Clin Oncol 20:7-8, 2002 5. Iyer L, Das S, Janisch L, et al: UGT1A1*28 polymorphism as a determinant of irinotecan disposition and toxicity. Pharmacogenomics J 2:43-47, 2002[CrossRef][Medline]
6. Ando Y, Saka H, Asai G, et al: UGT1A1 genotypes and glucuronidation of SN-38, the active metabolite of irinotecan. Ann Oncol 9:845-847, 1998 7. Iyer L, King CD, Whitington PF, et al: Genetic predisposition to the metabolism of irinotecan (CPT-11): Role of uridine diphosphate glucuronosyltransferase isoform 1A1 in the glucuronidation of its active metabolite (SN-38) in human liver microsomes. J Clin Invest 101:847-854, 1998[Medline]
8. Wasserman E, Myara A, Lokiec F, et al: Severe CPT-11 toxicity in patients with Gilbert’s syndrome: Two case reports. Ann Oncol 8:1049-1051, 1997 9. Freyer G, Rougier P, Bugat R, et al: Prognostic factors for tumour response, progression-free survival and toxicity in metastatic colorectal cancer patients given irinotecan (CPT-11) as second-line chemotherapy after 5FU failure: CPT-11 F205, F220, F221 and V222 study groups. Br J Cancer 83:431-437, 2000[CrossRef][Medline] 10. Wasserman E, Myara A, Paumier D, et al: Baseline bilirubin (Bil) and its transient early increase predicts likelihood of severe neutropenia and diarrhea in CPT-11-based chemotherapy. Proc Am Soc Clin Oncol 16: 1997 (abstr 767)
11. Fuchs CS, Moore MR, Harker G, et al: Phase III comparison of two irinotecan dosing regimens in second-line therapy of metastatic colorectal cancer. J Clin Oncol 21:807-814, 2003 12. Kaplan E, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958[CrossRef] 13. Cox D: Regression models and life tables. J R Stat Soc B 34:187-220, 1972 14. National Cancer Institute: Guidelines for the reporting of adverse drug reactions. Bethesda, MD, Division of Cancer Treatment, National Cancer Institute, 1999
15. Carmel R, Wong ET, Weiner JM, et al: Racial differences in serum total bilirubin levels in health and in disease (pernicious anemia). JAMA 253:3416-3418, 1985
16. Rosenthal P, Pincus M, Fink D: Sex- and age-related differences in bilirubin concentrations in serum. Clin Chem 30:1380-1382, 1984 17. Schaaf L, Rothenberg M, Hammond L, et al: Phase I trial of weekly irinotecan (CPT-11) in patients with hepatic dysfunction. Proc Am Soc Clin Oncol 20:101a, 2001 (abstr 402)
18. Raymond E, Boige V, Faivre S, et al: Dosage adjustment and pharmacokinetic profile of irinotecan in cancer patients with hepatic dysfunction. J Clin Oncol 20:4303-4312, 2002 19. Innocenti F, Undevia SD, Iyer L, et al: UGT1A1*28 polymorphism is a predictor of neutropenia in irinotecan chemotherapy. Proc Am Soc Clin Oncol 22:124, 2003 (abstr 495)
20. Conti JA, Kemeny NE, Saltz LB, et al: Irinotecan is an active agent in untreated patients with metastatic colorectal cancer. J Clin Oncol 14:709-715, 1996
21. Saltz LB, Douillard JY, Pirotta N, et al: Irinotecan plus fluorouracil/leucovorin for metastatic colorectal cancer: A new survival standard. Oncologist 6:81-91, 2001 22. 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] 23. Kerr D: Capecitabine/irinotecan in colorectal cancer: European early-phase data and planned trials. Oncology (Huntingt) 16:12-15, 2002 Submitted October 8, 2003; accepted January 16, 2004.
Related Correspondence
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2004 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
