Plasma 5-fluorouracil (5-FU) concentrations vary greatly between individuals who have received standard dosage. Pharmacokinetic adjusted doses have been hypothesized to overcome the possibility of potential toxicity and ineffectiveness related to inappropriate plasma levels of 5-FU. In this study, we prospectively investigated the clinical benefit and toxicity of 5-FU in relation to its pharmacokinetic properties.
Pharmacokinetics, effectiveness, and toxicity of 5-FU were investigated in 101 patients. The 5-FU pharmacokinetics were measured on day 2 of chemotherapy infusions. Clinicodemographic characteristics are outlined.
All 101 patients who received adjuvant chemotherapy were alive at the end of a median 45 months of the follow-up period. At least one grade 1 adverse event (AE) was observed in 69.3% of the patients and grade two AEs were observed in 10.1% of the patients. The 5-FU levels ranged between 103 and 4311 µg/L and area under the curve (AUC) measurements ranged between 4.5 and 189.7 mg min/L. Pharmacokinetic measurements were not significantly correlated with clinical efficacy (log-rank p = 0.21). However, higher AUC levels were positively correlated with toxicity (p = 0.02) and with the severity of adverse events. The risks of mucositis (odds ratio [OR] 1.45; p = 0.042) and neurotoxicity (OR 2.01; p = 0.009) were significantly increased in a logistic regression model.
There is no clear evidence that increased plasma levels or pharmacokinetic adjusted doses of 5-FU were related to better efficacy. However, toxicity might be closely associated with increased plasma levels of 5-FU. Toxicities can be deferred via dose adjustments without any expense in efficacy.
Post author correction
Article Type: ORIGINAL RESEARCH ARTICLE
AuthorsEce Esin, Tugba Akin Telli, Deniz Yuce, Suayib Yalcin
- • Accepted on 22/05/2017
- • Available online on 08/07/2017
This article is available as full text PDF.
Prognosis of advanced colorectal cancer has dramatically changed in recent years with the anti-vascular endothelial growth factor and anti-epidermal growth factor receptor agents, with the median overall survival now exceeding 33 months (1, 2). What has not changed in a decade in the treatment of colorectal cancer is the backbone with fluorouracil compounds (3). Infusional 5-fluorouracil (5-FU) regimens (De Gramont and FOLFOX4 or FOLFOX6) are commonly used for colorectal cancer in which fluorouracil calculations based on body surface area (BSA). However, 5-FU has a narrow therapeutic index and high variance (4). These individual differences in plasma levels depend on genetic diversity in the metabolic enzymes of 5-FU (i.e., dihydropyrimidine dehydrogenase [DPD]), as well as kinetic changes related to age, sex, body weight, and dietary differences (5, 6). Previous reports showed that inappropriate dosing is related to increased toxicity and increased rates of unresponsiveness (5, 7). Pharmacokinetic (PK) adjusted doses may overcome the problems of variability (8-9-10-11).
Area under the target curve of 20-25 mg × h/L was proposed as the valid 5-FU level to achieve clinically effective and less toxic chemotherapy cycles (12-13-14). However, with BSA-based dosing, only 20%-30% of patients received the valid area under the curve (AUC) dose, whereas nearly 50% of patients were underdosed and 20% were overdosed (8). Gamelin et al suggested increasing the overall response rate from 18% to 33% by PK dosing (14).
In this study, we prospectively investigated the clinical benefit and toxicity of 5-FU in relation to pharmacokinetic properties in advanced colorectal cancer and in the adjuvant setting.
Patients from an outpatient clinic who were diagnosed pathologically with colorectal cancer, who were 18 and older and were treated either in an adjuvant or metastatic setting, were eligible for this study. The Response Evaluation Criteria in Solid Tumors (RECIST) group on computed tomography (CT) were used for the assessment of tumor responses to chemotherapy. Ascites and pleural effusion were not considered as measurable diseases. Ineligibility criteria included poor performance status (Eastern Cooperative Oncology Group Performance Status >2), less than 3 months of life expectancy, pregnancy, lactation, other malignancies except basal cell skin carcinoma, abnormal liver function tests (serum creatinine >1.5 mg/dL, total bilirubin × 1.5 times upper limit of normal [ULN], and aspartate and alanine aminotransferase >2.5 × ULN [or >5 × ULN in the presence of liver metastases]), and inadequate hematopoietic function (platelet count <100,000/L and neutrophils <1.5 × 109/L). A written informed consent was obtained from all patients. The local ethics committee approved the protocol of the study. Informed consent was obtained from participants.
The patients were treated with mFOLFOX6, FOLFOX4 with or without biological agents, or infusional FUFA as determined by the primary physician. The chemotherapy schemes, doses, and application duration are outlined in
|Extended ras status|
|Primary tumor site|
|Modified de Gramont||1||1.0|
|FOLFOX-6 + cetuximab||3||3.0|
|FOLFOX-6 + bevacizumab||18||17.8|
|FOLFIRI + bevacizumab||31||30.7|
|FOLFIRI + cetuximab||11||10.9|
The fluorouracil pumps were prefilled in the central pharmacy in preordered BSA calculated doses (5-FU dose intensities per cycle according to different treatment schemes are outlined in
Follow-up, assessment of response, and assessment of toxicity
The treatment continued until progressive disease, intolerable toxicity, or withdrawal of consent. Patients were examined biweekly. No pharmacokinetic dose adjustment was done in the following chemotherapy cycles. Toxicities were assessed using the National Cancer Institute Common Terminology Criteria for Adverse Events v3.0 and were recorded in each cycle and afterwards. New recordings were taken if drug dose was changed. Blood tests were done in each cycle of chemotherapy (hemogram, liver function tests, and creatinine). Additionally, documentation of comorbid diseases and concomitant medications were recorded at each cycle. Clinical benefits were tested after 3 and 6 months with CT scans, and were investigated according to RECIST criteria (E.E. and S.Y.). Clinical outcomes were measured at the end of the follow-up.
This study was designed as a prospective cohort study. The primary efficacy endpoint was 3-month tumor response and 5-FU pharmacokinetic dose relationship, whereas the secondary efficacy endpoint was 6-month tumor response and 5-FU pharmacokinetic dose relationship. Any adverse events were analyzed at the secondary efficacy endpoint.
An objective response rate of at least 18% was anticipated with combination chemotherapy in accordance with the results of previous trials. Assuming a 5% type I error rate and 80% power, 101 patients were calculated as required sample size. With the assumption that 10% of patients might be excluded, a total of 111 patients were initially planned to be recruited to the study. Chi-square test was used for categorical variables and Mann-Whitney test for continuous variables. The confidence interval (CI) was accepted as 95% throughout the analyses. Kaplan-Meier method and log-rank test were used to calculate the survival data and Cox regression model was performed for multivariate analysis. All statistical analyses were performed using by IBM SPSS (Chicago, IL) software v22.
In the period between January 2013 and March 2013, 115 eligible patients were selected for the study. Three patients refused to receive chemotherapy, 1 patient withdrew consent, 5 patients continued the chemotherapy in another cancer center, and 3 patients discontinued before follow-up. Blood samples for pharmacokinetic analyses were taken from 103 patients, but in 2 of the cases the results were inappropriate due to errors in sampling and analysis. At the end, 101 patients were included in the study (Consolidated Standards of Reporting Trials diagram;
The most common sites of disease were rectum (26.7%), sigmoid colon (25.7%), recto-sigmoid area tumors (17.8%), and ascending colon (11.9%). In general, 70.3% were left-sided tumors and 19.8% were right-sided tumors. Extended ras testing was conducted in 67 cases. Ras wild-type tumors were present in 31.7% of patients, and the most frequent ras mutation was exon 2 codon-12 mutation in 26.7% of cases. Of all the participants in the study, 69 were treated for metastatic disease and 32 were treated in adjuvant setting (2 cases for stage II disease, 30 cases for stage III disease). The most frequent metastatic sites were liver (40.6%), lung (21.7%), and lymph nodes (34.8%). A total of 36 patients received at least one prior line of chemotherapy. Pharmacokinetic studies were conducted irrespective of the types of chemotherapy that patients received. Hence, the most common 5-FU-including regimens were distributed as follows: FOLFIRI + bevacizumab (30.7%), mFOLFOX6 + bevacizumab (17.8%), FOLFIRI + cetuximab (10.9%). The details are outlined in
Treatment schedules and cumulative doses
|Dose per infusion/m2||Infusion duration||Cumulative dose/cycle||Cumulative dose/treatment duration in 6 months|
|Infusional 5-FUFA (23, 24)||400 mg/m2 IV bolus on day 1||Bolus||2,800 mg/m2||33,600 mg/m2|
|2,400 mg/m2 IV||Over 46 hours|
|FOLFOX4 (3)||400 mg/m2||Over 2 hours on days 1 and 2||2,000 mg/m2||24,000 mg/m2|
|600 mg/m2 on days 1 and 2||Over 22 hours on days 1 and 2|
|mFOLFOX6 (23, 24)||400 mg/m2 IV bolus on day 1||Bolus||2,800 mg/m2||33,600 mg/m2|
|2,400 mg/m2 IV||Over 46 hours|
|FOLFIRI (25)||400 mg/m2 IV bolus on day 1||Bolus||2,800 mg/m2||33,600 mg/m2|
|2,400 mg/m2 IV||Over 46 hours|
The 5-FU levels ranged between 103 and 4,311 µg/L and the AUC measurements ranged between 4.5 and 189.7 mg × min/L. The correlation analyses between age, sex, body weight, and 5-FU pharmacokinetics showed that there is a significant relationship between the body weight and 5-FU plasma levels. Target AUC (20-25 mg × h/L) was achieved in 23.8% of cases. In metastatic patients, the rate of reaching the target value was 27.5%, while it was only 3% in the adjuvant setting. Body surface area-based calculations showed that of all the participants, 45.5% of patients were treated with lower than target PK doses, whereas 36.2% of the metastatic patients and 71.9% of the participants in the adjuvant setting were underdosed. Of all the participants, 30.7% of patients were treated with higher than target PK doses, whereas 36.2% of the metastatic patients and 25% of the participants in the adjuvant setting were overdosed. The distribution of AUC levels and the comparison of AUC levels of the respondent and nonrespondent group are presented in
Boxplot graphic of area under the curve (AUC) distribution in responders and nonresponders and according to presence of toxicity.
Median follow-up time for patients who received adjuvant treatment was 45 months (10-46). In the follow-up, metastasis occurred in 2 cases in the liver, and in 1 case in the lungs. All of the patients except 1 with liver and 1 with lung metastasis were alive at the end of the follow-up period in the adjuvant treatment group. All 3 progressions occurred in patients with low AUC levels (<25 mg × min/L). Median overall survival was not reached for patients in the adjuvant treatment group. At the end of the follow-up period, 96.9% of the patients were alive and 90.6% of the patients were disease-free.
Of patients with metastatic disease, at the end of the 3-month follow-up period, 28 of them had partial remission, 30 of them had stable disease (SD), and 11 had progressive disease (PD). Three-month clinical benefit rate (PR + SD) in metastatic patients was calculated as 84% and 6-month clinical benefit rate dropped to 36.2%. The pharmacokinetic analysis showed that the AUC levels were not correlated with either 3-month (p = 0.349) or 6-month clinical benefit rate (p = 0.839;
The relationship between the area under the curve (AUC) levels and overall survival in metastatic patients.
Toxicities according to pharmacokinetics. AUC = area under the curve.
In 24 patients, the cycle of therapies ended without any toxicity (24.8%); however, 75.2% of patients experienced at least one adverse event. The incidence rates of adverse events are outlined in
Incidence rates and grades of toxicities
|All grades, n (%)||Grade 3/4, n (%)|
|Any||76 (75.2)||8 (7.9)|
|Neutropenia||16 (15.8)||2 (2)|
|Mucositis||35 (34.7)||1 (1)|
|Emesis||48 (47.5)||2 (2)|
|Diarrhea||36 (35.6)||1 (1)|
|Hand foot syndrome||32 (31.7)||1 (1)|
|Ocular toxicity||16 (15.8)||0|
|Cardiovascular toxicity||3 (30)||1 (1)|
The rates of toxicity at each AUC range are illustrated in
Recent advances in colorectal treatment led to increased survival rates, while fluorouracil remained as an indispensable backbone of all chemotherapy regimens. Fluorouracil is used in BSA-related doses, which has some drawbacks. High interpatient variability, different metabolism characteristics of individuals, age, sex, and dietary influence on plasma levels all make 5-FU an unpredictable compound for implementation in chemotherapy (4, 6, 7, 10, 12, 16). Previous studies demonstrated that not only the exposure to 5-FU differs on an individual basis, and BSA-related dosing leads to suboptimal treatment, but also the toxicity of chemotherapies is related to inconstant 5-FU plasma levels (10, 13, 16). Hence, investigators concluded that BSA is not an appropriate metric for 5-FU. In order to overcome the variability issues, pharmacokinetics-adjusted doses were proposed instead of BSA-related dosing (4, 12, 14, 16, 18). In addition, in routine daily practice, clinicians confront the challenge of avoiding toxicity without sacrificing efficacy. The aim of this study was to investigate the efficacy and toxicity of 5-FU in relation to its pharmacokinetic properties in a cohort of colorectal cancer patients treated in adjuvant and metastatic settings prospectively.
The present trial demonstrated the treatment outcomes and toxicity of a divergent patient population. Overall, the rate of achieving target AUC level is 23.8% and the 5-FU levels ranged between 103 and 4,311 µg/L. It was demonstrated that 45.5% of patients were underdosed according to PK measurements. These findings were comparable with the results from previous studies (14, 19, 20). In the adjuvant treatment subgroup, the rate of achieving target AUC was as low as 3% and 71.9% of patients received lower than required doses. Of metastatic subgroup of patients, 30.7% of them exposed to lower than expected range of plasma levels of FU. These results reflect that BSA is not an appropriate and accurate metric for 5-FU in chemotherapy calculations.
At the end of the follow-up period, 96.9% of the patients were alive and 90.6% of the patients were disease-free in the adjuvant treatment group. Of the metastatic patients, 3-month clinical benefit rate (PR + SD) was calculated as 84% and 6-month clinical benefit rate as 36.2%. The pharmacokinetic analysis did not show a significant relationship between the AUC levels and clinical benefit rate. The median overall survival of metastatic patients was 45 months, which is among the highest survival rates of previous trials (8, 12, 17, 19, 21, 22). The lower rates of required target AUC levels and lower than required doses were not reflected in the clinical efficacy results of this study. The lower range of pharmacokinetic measured doses and lower 5-FU exposure did not result in worse outcomes, which contradicts the findings from previous studies (22). As indicated in a recent meta-analysis, much of the clinical benefit data came from studies that utilized outdated chemotherapy regimens (22). The present study included newer, more popular regimens like FOLFOX6 and cumulative 5-FU doses, which were similar between patients except in those who were treated with FOLFOX4 (26.7%). Besides, in the previous trials, the comparative analysis was done between heterogeneous patient groups, and the tests were administered using different measurement techniques (My5-FU immunoassay, high-performance liquid chromatography) (9, 11, 14, 18).
The next step of investigation in the present trial was the toxicity. Of the entire patient population, 75.2% experienced at least one adverse event. Furthermore, grade 3-4 toxicity was observed in 7.9% of cases. All of the grade 3-4 toxicities were observed in patients with high AUC values (>25 mg × h/L), and more side effects were experienced by patients who received higher than optimal doses. The analyses revealed that the rate of toxicities and the grade of adverse events were significantly related to overdosing. In particular, the incidence of mucositis and the risk of neurotoxicity were closely related to higher 5-FU exposure. The results of this study were in conjunction with previous clinical trials (5, 10, 12, 17). Given the previous research and the results from the present study, we can conclude that unexpected toxicity leads to higher than optimally desired 5-FU exposure.
A limitation of this study is that there are no comparative analyses available. The efficacy and toxicity were measured in a cohort of patients within a limited time period, and no adjustments on doses of 5-FU were done according to pharmacokinetic results. In addition, the patient population was diverse in characteristics, tumor types, and chemotherapy schemes, limiting the strength of the conclusions. The cumulative 5-FU doses could complicate the results of the benefit and toxicity but the schemes that are used mainly included 5-FU doses of updated chemotherapy regimens and cumulative doses were within the accepted optimal ranges (23-24-25). Furthermore, although the regimens were different from each other, the blood samples were taken in the first cycle and side effects were investigated for that cycle of treatment. Hence, the dose intensity for each cycle is comparable between the schemes and the patients. The results of the randomized controlled studies do not always reflect real-world data. The heterogeneity of cohort studies and the parallel results of analyses of the previous studies in this cohort might refer to the power of this study.
The elimination of 5-FU is one of the denominators of the pharmacokinetics of this drug and the clearance of 5-FU is mainly affected by renal excretion and hepatic metabolic degradation. The main rate-limiting enzymes in the clearance of 5-FU are DPD and thymidylate synthetase, the 2 well-studied predictive factors for toxicity. The activity of these enzymes varies individually, arising from genetic polymorphisms. Although rare, genetic alterations result in fatal toxicities. The 2A, 13, and 9B are the most commonly reported DPD variants responsible for the detected DPD-related toxicities (12, 26, 27). In addition to DPD, high-risk polymorphisms in the thymidylate synthetase gene may be associated with 1.4- to 2.4-fold increase in the risk of severe 5-FU toxicity (28, 29). However, the clinical evidence regarding thymidylate synthetase gene polymorphisms is less clear than DPD polymorphisms. A major issue is that well-known, commercially available high-risk alleles do not account for all cases of DPD and thymidylate synthetase gene polymorphisms. Although genotyping for risky alleles has the potential for identifying patients who are at risk of severe toxicity, reports showed that approximately half of the patients do not have an identifiable mutation, and therefore, the sensitivity of tests is limited (17, 27). Furthermore, genetics do not always account for phenotype and some of the patients harboring mutations or variant polymorphisms show signs of toxicity. Combined use of therapeutic drug monitoring and genotyping of drug metabolic capacity can be considered as the most sophisticated way of individualization of drugs in order to achieve optimal clinical benefit rates with minimum toxicity. The results of this study could have demonstrated the link between toxicity and 5-FU doses better if genotyping or phenotyping of DPD or thymidylate synthase could have been done.
For routine medical practice, the main barrier in the choice of optimal chemotherapy regimen and in the utilization of optimal doses is the anticipated toxicity. In order to avoid adverse events, lesser than optimal doses were generally used or dose decrements were done in the forfeit of lesser clinical efficacy and shorter survival. On the other hand, the results of the present study demonstrate that the doses based on BSA can be decreased in order to prevent toxicity within a safe limit while not compromising efficacy. Fluorouracil is the one of the most commonly used compounds worldwide; however, pharmacokinetic measurement techniques and equipment are not as widely available as the 5-FU itself. Thus, although BSA is not a gold standard metric for calculations of 5-FU, it is easy, practical, and acceptable. When available, pharmacokinetic measurements should be applied and adjustments of doses should be done on these measurements. In other circumstances, BSA-based calculations offer safe and efficient treatment options for patients.
Lenz H Niedzwiecki D Innocenti F al. CALGB/SWOG 80405: phase III trial of irinotecan/5-FU/leucovorin (FOLFIRI) or oxaliplatin/5-FU/leucovorin (MFOLFOX6) with bevacizumab (BV) or cetuximab (CET) for patients (PTS) with expanded ras analyses untreated metastatic adenocarcinoma of the colon or rectum (MCRC). 2014 25 suppl 4 mdu438. 13.
- Esin, Ece [PubMed] [Google Scholar] 1
- Akin Telli, Tugba [PubMed] [Google Scholar] 2
- Yuce, Deniz [PubMed] [Google Scholar] 3
- Yalcin, Suayib [PubMed] [Google Scholar] 1, * Corresponding Author (firstname.lastname@example.org)
Department of Medical Oncology, Hacettepe University Cancer Institute, Ankara - Turkey
Department of Medical Oncology, Marmara University, Istanbul - Turkey
Department of Prevantive Oncology, Hacettepe University Cancer Institute, Ankara - Turkey