Department of Clinical Pharmacology, University of Oxford, Oxford, UK
* Current address: Department of clinical oncology, University of Hong Kong, Hong Kong
Colorectal cancer (CRC) is one of the leading malignancies in the Western World, responsible for 400 000 deaths each year [1
, 2
]. Despite radical surgery, more than half of CRC patients develop metastases, and palliative chemotherapy has been utilised to alleviate symptoms and prolong survival [3
, 4
]. The availability of newer and more efficacious agents over the last few years has seen an improvement in median overall survival in these patients, particularly when the three agents, 5-fluorouracil (5-FU), irinotecan and oxaliplatin, are all used at some stage in the clinical course [5
10
].
Nonetheless, the success of chemotherapy is still far from perfect, as tumour response is only observed in 50% of patients and at the expense of toxicities, which in most subjects are mild to moderate but for other patients may be unpredictable and severe. In this regard, cytotoxic treatment is unique in that it tends to operate in a narrow therapeutic window, the result of an oncological dogma that dictates the use of these drugs at near the maximum tolerated dose. Irinotecan, a topoisomerase I inhibitor with useful activity in CRC, carries risks of significant morbidities, including a gastrointestinal syndrome characterised by acute and delayed diarrhoea, neutropenia and a vascular syndrome, and fatalities associated with these toxicities have been reported [11
13
]. Moreover, chemotherapy drug dosage is routinely calculated according to the patient's body surface area, a variable of the subject's weight and height. However, body surface area has been shown to be an imprecise method of dosage estimation for many chemotherapeutic drugs [14
, 15
], and in the case of irinotecan, it does not accurately predict irinotecan clearance or SN-38 pharmacokinetics, and does not contribute to reducing kinetic variability [16
].
Equally unpredictable in CRC chemotherapy is the efficacy for an individual patient. A blanket policy of palliative chemotherapy results in a significant proportion of patients being subjected to potential side-effects without any significant gain in symptom relief and survival. Besides, the increasing diversity of drugs and the expanding repertoire of combinations and schedules, though an advancement in our armamentarium, often complicate the clinical situation, as individual patients may undergo an ineffective treatment or one that is overly toxic before being commenced on a drug with a more favourable therapeutic ratio. More importantly, although it is now known that the use of all three agents, 5-FU, irinotecan and oxaliplatin, may extend the median survival further, there is still a lack of consensus on the correct sequence in which these agents should be offered [9, 10
]. Such knowledge, specific for an individual patient, if known, may improve the clinical outcome and relieve many patients of unnecessary toxicities. That such a need for individualisation or tailoring of treatment is so important can be seen, as it is being hailed by some authors as the Holy Grail of medical practice [17
].
Until therapy tailoring becomes a reality, there are several ways in which maximisation of efficacy and minimisation of toxicity may be effected. Most clinicians rely on data generated from large population cohorts when patients with the same disease are being treated with a similar therapy. Various parameters will be analysed by multivariate methods in order to distill out their individual contribution to the outcome, be that efficacy or toxicity. These parameters are numerous, and range from patient factors to multiple clinico-pathological variables; when factored together, they improve prognostication and prediction. Indeed, for palliative chemotherapy in advanced CRC, individualisation has been helped by data generated by large pooled studies in the case of 5-FU-based treatment, and patients can be stratified into risks groups with distinct median survival times [18]. In a similar manner, subgroups of patients with distinct response rates, progression-free survival times and rates of toxicities can be identified by consideration of a number of clinical factors when irinotecan is used in the second-line setting [19
]. In the USA, where the IFL regimen (bolus irinotecan plus 5-FU and leucovorin) used to dominate as the standard first-line regimen, elevation in bilirubin level was found to be associated with grade 3 or 4 neutropenia [20
].
For many European countries and other parts of the world where irinotecan plus fluorouracil and leucovorin are commonly given as an infusional regimen, there are very limited data on which outcome and toxicities can be predicted. It is therefore timely that Mitry et al. [21], in this issue of Annals of Oncology, present their analysis of pooled individual data for patients included in two phase III trials: the V302, in which irinotecan was compared in the second-line setting with an infusional 5-FU in the form of either LV5FU2 or AIO regimen, and the V303 study, where irinotecan plus the same LV5FU2 or AIO regimen was compared with those infusional regimens alone in the first-line setting [7
, 21
, 22
]. The data for patients in the V302 trial treated with the Lokich regimen were not included in the analysis in order to maintain homogeneity of the 5-FU alone arm across these two studies. Patients were stratified according to the treatment received into irinotecan-based or otherwise. The authors begin by analysing a large number of potential clinico-pathological factors with univariate comparison. On multivariate analysis, preserved performance status, lack of significant weight loss and treatment with irinotecan were found to be associated with a better progression-free survival. These three factors, along with a low serum alkaline phosphatase level and the presence of two or fewer sites of metastases, predicted a better overall survival [21
].
The merit of the study by Mitry et al. is that they are drawing reliable data from two well-conducted phase III trials in which the control arm for comparison is clinically relevant, as these regimens are still in active use in the clinic. The reasonably large number of patients included in the present study also helps to strengthen the findings and conclusions. Noteworthy, however, is that Mitry and colleagues, in combining these two studies, are mixing data for single-agent irinotecan with those for infusional irinotecan plus 5-FU and leucovorin. Such heterogeneity, together with the pooling of first- and second-line data, makes interpretation problematic, as the clinical and physiological conditions of patients in first- and second-line settings are often different and the factors governing the outcome may not be same for single agent and combination regimens. Notwithstanding such shortcomings, these predictive factors are useful in selecting patients who may respond to irinotecan treatment and hence maximise the chance of a better survival with this drug.
The reality remains that, to date, clinical decisions are still largely dependent on average data from a population rather than on the specific conditions of the patient. Indeed, truly individualised treatment should rely on more personalised data, and recent work on pharmacokinetics, pharmacodynamics, biological and molecular predictive markers and pharmacogenomics are gaining headway in this regard. Given the potentially severe toxicities of irinotecan and its widening role in CRC, it represents an excellent model for studies in these areas. Plasma concentration of irinotecan and its active and inactive metabolites, SN-38 and SN-38G, represented by the area under the curve, have been associated with various toxicities [23]. Although these pharmacokinetic parameters show marked inter-individual variability [24
] and hence cannot be used as the sole determinant in therapy decision, they provide important information about the kinetics of drug metabolism that may be useful in dosage optimisation [25
].
Similarly, there has been significant increase in the knowledge of many biological and molecular markers that may predict efficacy or toxicity. In CRC, where 5-FU-based treatment predominates, tumoural expression of various proteins and enzymes involved in the metabolism of 5-FU, for example, thymidylate synthase (TS), has been shown to correlate with outcome [2628
]. In addition, somatic alterations, for instance, loss of heterogeneity, have been shown to predict survival in CRC patients receiving adjuvant fluorouracil therapy [29
]. For oxaliplatin, tumoural expression of ERCC1 protein, an integral part of the nucleotide excision repair system (NER) responsible for repair of platinum-induced DNA damage, has been reported to correlate with outcome in advanced CRC patients treated with oxaliplatin and resistance towards this novel platinum in colon cancer cell lines study [30
, 31
]. Other molecular factors may predict for treatment morbidity and in patients receiving 5-FU-based therapy, low activity of dihydropyrimidine dehydrogenase, a key enzyme responsible for its catabolism, has been shown to correlate with toxicity [32
34
].
Factors governing these inter-individual variations in pharmacokineticpharmacodynamic parameters and expressions of the many metabolic proteins or targets of chemotherapeutic agents in normal tissues and tumours are numerous and diverse, but the most important mechanism underlining all these different phenotypes must lie in the genetic variations between patients. With the explosion of knowledge obtained recently regarding the human genome, the study of genomics is rapidly transforming the discipline of pharmacogenetics, and trials on genotyping specific phenotypes have been reported. For irinotecan, the inactivation of the active metabolite SN-38 is carried out by the uridine diphosphate glucuronosyltransferase 1A1 (UGT1A1). Germ-line polymorphisms involving UGT1A1 have been reported and the TA insertion in the TATA sequence of the promotor region of UGT1A1 results in a variant allele (TA)7TAA instead of the wild-type (TA)6TAA, leading to reduced UGT1A1 expression and activity [35]. Subsequent clinical data demonstrated that heightened toxicities were more likely to be observed in patients harbouring the variant seven TA repeat allele [36
, 37
]. Another example of the potential of pharmacogenomics is the correlation shown between a polymorphic tandem 28 base-pair repeat in the TS promoter region and the expression of TS [38
]. This genotype was found to correlate with outcome in patients treated with 5-FU in both adjuvant and palliative settings [39
41
], and may also be predictive of tumour down-staging in rectal cancer patients treated with preoperative chemoradiotherapy [42
]. The above-mentioned NER protein, ERCC1, also has numerous single nucleotide polymorphisms (SNPs), and two of these genotypes were recently observed to correlate with outcome after treatment with oxaliplatin-based therapy [43
].
The same can be said about epidermal growth factor receptor [44], xeroderma pigmentosum group D [45
] and glutathione S-transferase P1 [46
]. More recently, germ-line polymorphism in TYMS (the gene encoding TS) [47
] was shown to correlate with response to 5-FU-based therapy, whereas germ-line polymorphisms of orotate phosphoribosyl transferase and TS tandem repeat were demonstrated to correlate with toxicity of 5-FU therapy [48
]. There are also early data suggesting that normal tissue mRNA levels of ERCC1 and TS may predict pelvic relapse in patients with rectal cancer treated with chemoradiation [49
]. Moreover, with modern high-throughput technology and the knowledge of the human genome, DNA microarray has been used to analyse thousands of genes at the same time, and data showing a correlation between basal gene expression profiles and apoptosis induced by various chemotherapy in colon cancer cell lines demonstrate that such technology can be used to predict and distinguish response to multiple chemotherapeutic agents [50
].
Nonetheless, despite the excitement of these latest advances, to correctly read and interpret the pharmacogenomic data, well-conducted clinical studies with comprehensively and accurately recorded data like the one by Mitry and colleagues are of paramount importance. One such trial is the upcoming QUASAR II study, in which CRC patients requiring adjuvant chemotherapy will be randomised into one of three arms: 5-FU plus leucovorin, capecitabine plus irinotecan, or capecitabine plus irinotecan and avastin. Corollary population pharmacokinetic studies will be performed and prospective collection of patients genomic materials will be undertaken and analyses to determine the population distribution of SNPs in enzymes involved in the metabolism of 5-FU, capecitabine and irinotecan will be carried out and correlated with the clinical outcome of toxicity and recurrence rates.
Taken together, the recent breakthrough in genomic medicine does offer oncologists the possibility of individualised treatment, and to make this a reality, future trials should incorporate parallel studies in pharmacokineticspharmacodynamics, molecular determinants and pharmacogenomics. Together with the rapid improvement in technology and data analysis, it is highly probable that cancer patients in the near future may indeed be presented with the Holy Grail: tailored treatment with high efficacy and low toxicity.
References
1. Greenlee RT, Hill-Harmon MB, Murray T et al. Cancer Statistics 2001. CA Cancer J Clin 2001; 51: 1536.
2. Midgley R, Kerr D. Conventional cytotoxic and novel therapeutic concepts in colorectal cancer. Expert Opin Investig Drugs 2001; 10: 10111019.[ISI][Medline]
3. Scheithauer W, Rosen H, Kornek GV et al. Randomized comparison of combination chemotherapy plus supportive care or supportive care alone in patients with metastatic colorectal cancer. BMJ 1993; 306: 752755.[ISI][Medline]
4. Glimelius B, Hoffman K, Graf W et al. Quality of life during chemotherapy in patients with symptomatic advanced colorectal cancer. Cancer 1994; 73: 556562.[ISI][Medline]
5. de Gramont A, Figer AA, Seymore M et al. Leucovorin and fluorouracil with or without oxaliplatin as first line treatment in advanced colorectal cancer. J Clin Oncol 2000; 18: 29382947.
6. Saltz LB, Cox JV, Blanke C et al. Irinotecan plus fluorouracil and leucovorin for metastatic colorectal cancer. Irinotecan Study Group. N Engl J Med 2000; 343: 905914.
7. Douillard JY, Cunningham D, Roth AD et al. Irinotecan combined with fluorouracil compared with fluorouracil as first line treatment for metastatic colorectal cancer: a multicentre randomized trial. Lancet 2000; 355: 10411047.[CrossRef][ISI][Medline]
8. Goldberg RM, Sargent DJ, Morton RF et al. A randomized trial of fluorouracil plus leucovorin, irinotecan and oxaliplatin combinations in patients with previously untreated metastatic colorectal cancer. J Clin Oncol 2004; 22: 2330.
9. Tournigand C, Andre T, Achille E et al. FOLFIRI followed by FOLFOX6 or the reverse sequence in advanced colorectal cancer: a randomized GERCOR study. J Clin Oncol 2004; 22: 229237.
10. Grothey A, Sargent D, Goldberg RM et al. Survival of patients with advanced colorectal cancer improves with the availability of fluorouracilleucovorin, irinotecan, and oxaliplatin in the course of treatment. J Clin Oncol 2004; 22: 12091214.
11. Gupta E, Ratain MJ. Camptothecin analogues: topotecan and irinotecan. In Grochow LB, Ames M (eds): A Clinician's Guide to Chemotherapy Pharmacokinetics and Pharmacodynamics. Baltimore (MD): Williams & Wilkins 1998; 435457.
12. Ratain MJ. Editorial: Irinotecan dosing: Does the CPT in CPT-11 stand for "can't predict toxicity"? J Clin Oncol 2002; 20: 78.
13. Rothenberg ML, Meropol NJ, Poplin EA et al. Mortality associated with irinotecan plus bolus fluorouracil/leucovorin: summary findings of an independent panel. J Clin Oncol 2001; 19: 38013807.
14. Gurney HP, Ackland S, Gebski V et al. Factors affecting epirubicin pharmacokinetics and toxicity: evidence against using body surface area for dose calculation. J Clin Oncol 1998; 16: 22992304.[Abstract]
15. Baker SD, Verweij J, Rowinsky EK et al. Role of body surface area in dosing of investigational anticancer agents in adults, 19912001. J Natl Cancer Inst 2002; 94: 18831888.
16. Mathijssen RHJ, Verweij J, de Jonge MJA et al. Impact of body-size measures on irinotecan clearance: alternative dosing recommendations. J Clin Oncol 2002; 20: 8187.
17. Licinio J, Wong ML. Preface. In Licinio J, Wong ML (eds): Pharmacogenomics: The search for individualized therapies. Germany: Wiley-VCH: Weinheim 2002; pp. vii.
18. Kohne CH, Cunningham D, Di Costanzo F et al. Clinical determinants of survival in patients with 5-fluorouracil-based treatment for metastatic colorectal cancer: result of a multivariate analysis of 3825 patients. Ann Oncol 2002; 13: 308317.
19. 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. Br J Cancer 2000; 83: 431437.[CrossRef][ISI][Medline]
20. Knight RD, Miller L, Elfring G et al. Evaluation of age, gender, performance status (PS), and organ dysfunction as predictors of toxicity with first-line irinotecan (C), fluorouracil (F), leucovorin (L) therapy of metastatic colorectal cancer (MCRC). Proc Am Soc Clin Oncol 2001; 20: 134a (Abstr 534).
21. Mitry E, Douillard JY, van Cutsem E et al. Predictive factors of survival in patients with advanced colorectal cancer. An individual data analysis of 602 patients included in irinotecan phase III trials. Ann Oncol 2004; 15: 10131017.
22. Rougier P, van Cutsem E, Bajetta E et al. Randomized trial of irinotecan versus fluorouracil by continuous infusion after fluorouracil failure in patients with metastatic colorectal cancer. Lancet 1998; 352: 14071412.[CrossRef][ISI][Medline]
23. Pitot HC, Goldberg RM, Reid JM et al. Phase I dose-finding and pharmacokinetic trial of irinotecan hydrochloride (CPT-11) using a once-every-three-week dosing schedule for patients with advanced solid malignancy. Clin Cancer Res 2000; 6: 22362244.
24. Iyer L, King CD, Whittington PF et al. Genetic predisposition to the metabolism of irinotecan (CPT-11). J Clin Invest 1998; 101: 847854.
25. Sheiner LB. Population harmacokinetics/pharmacodynamics. Ann Rev Pharmacol Toxicol 1993; 32: 185200.[ISI]
26. Leichman CG, Lenz HJ, Leichman L et al. Quantitation of intratumoural thymidylate synthase expression predicts for disseminated colorectal cancer response and resistance to protracted-infusion fluorouracil and weekly leucovorin. J Clin Oncol 1997; 15: 32233229.[Abstract]
27. Aschele C, Bandelloni R, Cascinu S et al. Thymidylate synthase protein expression in colorectal cancer metastases predicts for clinical outcome to leucovorin-modulated bolus or infusional 5-fluorouracil but not methotrexate-modulated bolus 5-fluorouracil. Ann Oncol 2002; 13: 18821892.
28. Popat S, Matakidou A, Houlston RS. Thymidylate synthase and prognosis in colorectal cancer: a systematic review and meta-analysis. J Clin Oncol 2004; 22: 529536.
29. Elsaleh H, Joseph D, Grieu F et al. Association of tumour site and sex with survival benefit from adjuvant chemotherapy in colorectal cancer. Lancet 2000; 355: 17451750.[CrossRef][ISI][Medline]
30. Shirota Y, Stoehlmacher J, Brabender J et al. ERCC1 and thymidylate synthase mRNA levels predict survival for colorectal cancer patients receiving combination oxaliplatin and fluorouracil chemotherapy. J Clin Oncol 2001; 19: 42984304.
31. Arnould S, Hennebelle I, Canal P et al. Cellular determinants of oxaliplatin sensitivity in colon cancer cell lines. Eur J Cancer 2003; 39: 112119.[CrossRef][ISI][Medline]
32. Johnson MR, Hageboutros A, Wang K et al. Life-threatening toxicity in a dihydropyrimidine dehydrogenase-deficient patient after treatment with topical 5-fluorouracil. Clin Cancer Res 1999; 5: 20062011.
33. Milano G, Etienne MC, Pierrefite V et al. Dihydropyrimidine dehydrogenase deficiency and fluorouracil-related toxicity. Br J Cancer 1999; 79: 627630.[CrossRef][ISI][Medline]
34. Wei X. Molecular basis of the human dihydropyrimidine dehydrogenase deficiency and 5-fluorouracil toxicity. J Clin Invest 1996; 98: 610615.
35. Iyer L, Hall D, Das S et al. Phenotype-genotype correlation of in vitro SN-38 (active metabolite of irinotecan) and bilirubin glucuronidation in human liver tissue with UGT1A1 promotor polymorphism. Clin Pharmacol Ther 1999; 65: 576582.[ISI][Medline]
36. Iyer L, Das S, Janisch L et al. UGU1A1*28 polymorphism as a determinant of irinotecan disposition and toxicity. Pharmacogenomics J 2002; 2: 4347.[CrossRef][Medline]
37. Innocenti F, Undevia SD, Iyer L et al. Genetic variants in the UDP-glucuronosyltransferase 1A1 gene predict the risk of severe neutropenia of irinotecan. J Clin Oncol 2004; 22: 13821388.
38. Kawakami K, Omura K, Kanehira E et al. Polymorphic tandem repeats in the thymidylate synthase gene is associated with its protein expression in human gastrointestinal cancers. Anticancer Res 1999; 19: 32493252.[ISI][Medline]
39. Iacopetta B, Grieu F, Joseph D et al. A polymorphism in the enhancer region of the thymidylate synthase promoter influences the survival of colorectal cancer patients treated with 5-fluorouracil. Br J Cancer 2001; 85: 827830.[CrossRef][ISI][Medline]
40. Etienne MC, Chazal M, Laurent-Puig P et al. Prognostic value of tumoural thymidylate synthase and p53 in metastatic colorectal cancer patients receiving fluorouracil-based chemotherapy: phenotype and genotype analyses. J Clin Oncol 2002; 20: 28322843.
41. Marsh S, McKay JA, Cassidy J et al. Polymorphism in the thymidylarte synthase promnoter enhancer region in colorectal cancer. Int J Oncol 2001; 19: 383386.[ISI][Medline]
42. Villafranca E, Okruzhnov Y, Dominguez MA et al. Polymorphisms of the repeated sequences in the enhancer region of the thymidylate synthase gene promoter may predict downstaging after preoperative chemoradiation in rectal cancer. J Clin Oncol 2001; 19: 17791786.
43. Park D, Zhang W, Stoehlmacher J et al. ERCC1 polymorphism as a predictor for clinical outcome in advanced colorectal cancer patients treated with platinum-based chemotherapy. Clinical Advances in Hematology and Oncology 2003; 1: 120124.
44. Zhang W, Park DJ, Groshen S et al. A polymorphic dinucleotide repeat in intron 1 of EGFR (epidermal growth factor receptor) gene is associated with clinical response to platinum based chemotherapy in patients with advanced colorectal disease. Proc Am Soc Clin Oncol 2002; 21: (Abstr 533).
45. Park DJ, Stoehlmacher J, Zhang W et al. A xeroderma pigmentosum group D gene polymorphism predicts clinical outcome to platinum-based chemotherapy in patients with advanced colorectal cancer. Cancer Res 2001; 61: 86548658.
46. Stoehlmcher J, Park DJ, Zhang W et al. Association between glutathione S-transferase P1, T1 and M1 genetic polymorphism and survival of patients with metastatic colorectal cancer. J Natl Cancer Inst 2002; 94: 936942.
47. McLeod HL, Sargent DJ, Marsh S et al. Pharmacogenetic analysis of systemic toxicity and response after 5-fluorouracil (5FU)/CPT-11, 5FU/oxaliplatin (oxal), or CPT-11/oxal therapy for advanced colorectal cancer (CRC): results from an Intergroup tria. Proc Am Soc Clin Oncol 2003; 22: 253 (Abstr 1013).
48. Ichikawa W, Takahashi T, Nihei Z et al. Polymorphisms of orotate phosphoribosyl transferase (OPRT) gene and thymidylate synthase tandem repeat (TSTR) predict adverse events (AE) in colorectal cancer (CRC) patients treated with 5-fluorouracil (FU) plus leucovorin (LV). Proc Am Soc Clin Oncol 2003; 22: 265 (Abstr 1063).
49. Lenz HJ, Schneider S, Zhang W et al. Gene expression profile in normal tissue predicts pelvic recurrence in patients with rectal cancer treated with adjuvant chemoradiation therapy. Proc Am Soc Clin Oncol 2003; 22: 295 (Abstr 1185).
50. Mariadason JM, Arango D, Shi Q et al. Gene expression profiling-based prediction of response of colon carcinoma cells to 5-fluorouracil and camptothecin. Cancer Res 2003; 63: 87918812.