Research Highlights

Metastatic Colorectal Cancer: Full Speed on Treatment

Evaluation of: Zarate R, Rodriguez J, Bandres E et al.: Oxaliplatin, irinotecan and capecitabine as first-line therapy in metastatic colorectal cancer (mCRC): a dose-finding study and pharmacogenomic analysis. Br. J. Cancer 102, 987–994 (2010).

The treatment of metastatic colorectal cancer has seen major progress during the last decade. After a rather long period using 5-fluorouracil (5-FU) and leucovorin as treatment, two major trials clearly indicated a benefit of combination chemotherapy with both drugs [1,2]. Further progress has recently been obtained with biologically targeted treatment using monoclonal antibodies [3,4]. These effective chemotherapy regimens have changed our view on metastatic colorectal cancer. Until recently, treatment was considered palliative with survival prolongation, but it is now clear that a considerable proportion of previously inoperable patients can be switched to an operable status with the perspectives of curatively intended surgery.

The development of more effective regimens is therefore well motivated and should be given priority.

The paper by Zarate et al. clearly shows that the combination of the three effective drugs (oxaliplatin, irinotecan and capecitabine) is feasible [5]. The authors have performed a well-designed dose-finding study identifying the maximum tolerated dose for capecitabine with fixed doses of oxaliplatin and irinotecan. The dose limiting toxicities were diarrhea and neutropenia. The treatment was very effective, with a progression-free survival of 42 months in patients with favorable variables.

Treatment with more cytostatics usually results in more toxicity, as also seen in the study. Consequently, there is an obvious need for predictive markers for a rational selection of patients. The study suggests that the GSTPI-G genotype is of major importance, but it also shows that the conventional markers indicated by the Köhne index are of considerable value.

This is a small study, with analyses of 13 SNPs, the results of which should be interpreted with caution as to the effectiveness of the biomarkers described. We are still left with the question of optimal selection of patients for such hazardous treatment. Until we discover more and have performed large-scale studies, a possible option is to reserve the treatment with all three drugs for patients with the perspective of operation with curative intent.

Financial & competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.

Predictive Value of Germline Polymorphisms in Rectal Cancer

Evaluation of: Cecchin E, Agostini M, Pucciarelli S et al.: Tumor response is predicted by patient genetic profile in rectal cancer patients treated with neo-adjuvant chemo-radiotherapy. Pharmacogenomics J. DOI: 10.1038/tpj.2010.25 (2010) (Epub ahead of print).

Locally advanced rectal cancer represents a therapeutic challenge. Surgery alone is insufficient, and preoperative treatment is now considered standard in many countries. Radiation given before the operation is more effective than postoperative radiation, as shown by a large German trial [1]. The addition of chemotherapy improves efficacy but at the cost of higher toxicity [2]. This situation calls for the development of valid predictive markers to allow a rational treatment selection.

There is considerable literature on biomarkers in rectal cancer, but the results are divergent and, there is no single marker or marker profile generally accepted for clinical application.

One reason for this is that tumors are heterogeneous with respect many different biological characteristics, including biomarkers. Furthermore, biomarkers may vary over time with tumor progression. Germline polymorphism has the advantage of being stable and could therefore represent a source of markers of clinical interest, despite the fact that most SNPs seem to be of little or no biological importance.

The paper by Cecchin et al. indicates that the polymorphism holds important predictive information in locally advanced rectal cancer [3]. The authors analyzed 25 SNPs and later reduced the number to two (hOCC1 -1245>G and MTHFR -677C>T) by further statistical analyses with division of the patients into three groups. There was a highly significant difference in the rate of major response (complete pathological remission and microresidual tumor only) among the groups. The difference will probably be translated into a difference in long-term prognosis.

The results are based on retrospective analyses and should consequently be interpreted with caution, but the paper points to a target that should be pursued in well-designed prospective trials.

KRAS Mutation: Better Insight into the Gene Regulation

Evaluation of: Grazianio F, Canestrari E, Loupakis F et al.: Genetic modulation of the Let-7 microRNA binding to KRAS 3´-untranslated region and survival of metastatic colorectal cancer patients treated with salvage cetuximab-irinotecan. Pharmacogenomics J. DOI: 10.1038/tpj.2010.9 (2010) (Epub ahead of print).

The introduction of KRAS testing as a clinical tool for patient selection to anti-EGF receptor monoclonal antibodies has changed the therapeutic concept for metastatic colorectal cancer towards individualized treatment [1]. KRAS-mutant disease is not likely to respond to treatment – even in KRAS wild-type patients, the benefit is limited. The search for additional predictive markers should be encouraged and the complexity of KRAS regulation investigated.

miRNAs are able to inhibit RNA translation and are believed to play an important role in cancer. Specific miRNAs of the Let-7 family were recently discovered to influence the KRAS function and, furthermore, a genetic polymorphism (the LCS6 variant) was reported to alter the Let-7 binding to the KRAS mRNA. In brief, the different genotypes have been associated to different levels of KRAS activity, which indicates a functional influence of this germline polymorphism [2].

Graziano et al. have translated these findings into a clinical investigation. They have hereby added new aspects to the knowledge of KRAS mutations in colorectal cancer and opened up several important areas for investigation [3]. The study included 134 patients with irinotecan-resistant metastatic colorectal cancer treated with the combination of EGF receptor inhibitors cetuximab and irinotecan. The status of KRAS, BRAF and LCS6 was established. A different distribution of LCS6 G-allele carriers was reported according to KRAS mutational status, with the majority of G-alleles in KRAS-mutant tumors. It should be noted that BRAF-mutated tumors carried a LCS6 allele. The LCS6 G-allele was associated with worse progression-free survival rated and overall survival, which was confirmed in a multivariate analysis. Furthermore, a significant prognostic value of the polymorphism in KRAS-mutant disease was revealed. Whether Let-7 and this germline polymorphism influences the occurrence rate of KRAS mutations, suppresses constitutive KRAS activation or plays only a minor role in an increasingly complex picture is yet to be established.

Thus, the complexity of KRAS has increased with new interesting observations. Based on these results, we should await validation of the possible clinical importance but also explore the underlying mechanisms connecting the genetic variant to the EGF pathway.

Biomarkers For Cetuximab: A Complex Pattern and Need for Validation

Evaluation of: Tabernero J, Cervantes A, Rivera F et al.: Pharmacogenomic and pharmacoproteomic studies of cetuximab in metastatic colorectal cancer: biomarker analysis of a Phase I dose-escalation study. J. Clin. Oncol. 28(7), 1181–1189 (2010).

Treatment with monoclonal antibodies targeting the EGF receptor for metastatic colorectal cancer comes with a substantial degree of skin toxicity together with major financial cost. Consequently, the role of skin reactions, optimal treatment schedules and a better selection of patients are intensively debated aspects. A major step towards individualized treatment has been made with the discovery of a strong negative-predictive value of KRAS mutations for response to the EGF receptor inhibitors cetuximab and panitumumab [1,2]. Retrospective single-arm studies followed by the results of randomized trials have all contributed to the general conclusion that patients with KRAS mutations do not benefit from EGF receptor monoclonal antibodies. However, a major fraction of the KRAS wild-type patients also do not benefit from treatment. The search for additional predictive markers has been encouraged by the successful KRAS story, and both upstream markers by the levels of ligand expression [3] and downstream markers in terms of BRAF and PIK3CA mutations [4] were recently demonstrated to hold important predictive information.

The study by Tabernero et al. is an example of the increasing number of studies combining a clinical trial with a biomarker study in this setting [5]. The group investigated the possible optimization of the treatment schedule with biweekly instead of weekly cetuximab in 62 patients with metastatic colorectal cancer. Several other aspects were addressed in the primary study, namely monotherapy as first-line treatment, dose-escalation of cetuximab monotherapy followed by combination with folinic acid, 5-FU and irinotecan (FOLFIRI)/cetuximab, and the role of KRAS mutations. However, baseline and on-treatment tumor tissue, skin biopsies and blood samples were obtained from a subset of patients for pharmacogenomic and pharmacoproteomic analysis – a set-up requiring advanced technology and extended involvement from the patients.

The results confirmed the role of KRAS mutations in patient response, with particular emphasis on the monotherapy phase. It also showed a complex pattern of marker up- and down-regulation in the skin, but no significant differences according to treatment dose, which led the authors to conclude that the data support a functional equivalence of the two regimes. These data, combined with results of other studies, may change the clinical practice of treatment schedules with cetuximab [6,7].

A descriptive analysis of plasma marker pattern was performed, which included proteins directly and indirectly related to the EGF receptor system, together with already well-known potential markers, such as carcinoembryonic antigen, cancer antigen 125 and carbohydrate antigen 19–9. Although based on a small sample size, the study provides a basis for further analysis. This also applies to the results of the global gene-expression analysis in the repeated tumor biopsies. The complexity of genetic profiling has been underlined, and the distinction between KRAS-mutated and nonmutated patients was once again proven to be of major importance. Obviously, the time has come for larger scale investigations of candidate markers, such as those described, if a reliable clinical value is to be determined.