- © 2012 by American Society of Clinical Oncology
Linking Toxicity to the Target
In Journal of Clinical Oncology, Schneider et al1 report a study of taxane-induced neuropathy and clinical outcomes in a large randomized trial in early-stage breast cancer. The landmark E1199 trial established the optimal taxane and schedule (once-per-week paclitaxel or three-times-per-week docetaxel) to be delivered after a standard anthracycline regimen as adjuvant therapy in node-positive or high-risk node-negative breast cancer.2 The authors correlated the development of neuropathy within 7 months of random assignment among patients who received at least one dose of a taxane with recurrence-free survival, disease-free survival (DFS), and overall survival. No association was found between the development of grades 2 to 4 neuropathy and any of the clinical efficacy end points assessed.
Schneider et al1 should be commended for using the wealth of data extensively collected as part of clinical trials practice today. With all the efforts and resources required to both quantify and qualify all adverse events in clinical trials, it is encouraging to see productivity from such association studies. Furthermore, there was a rationale to attempt to correlate the development of neuropathy with clinical outcomes in E1199, because such an association was reported in another adjuvant breast trial (N9831).3 Differing results between the two studies may in part be explained by the fact that in the N9831 study, the investigators grouped all grades of neuropathy (grades 1 to 4) together, included only human epidermal growth factor receptor 2 (HER2) –positive breast cancer, and failed to account for other variables that could have also contributed to the development of neuropathy. As well, with the lack of good standardization for documenting neuropathy, factors as simple as reporting bias may also have played a role in the different results. Regardless, the trend for potential greater efficacy in the neuropathy cohort in the HER2-positive subgroup of E1199 (DFS: hazard ratio, 0.77; 95% CI, 0.52 to 1.12; P = .17), together with the results from the N9831 association study in the nontrastuzumab arm (DFS: HR, 0.64; 95% CI, 0.45 to 0.90; P = .01) and the knowledge of greater efficacy with the addition of taxanes after an anthracycline regimen in HER2-positive breast cancers,4 may lead one to hypothesize a possible correlation in this specific subtype of breast cancer, but this will require further validation.
But does it make any biologic sense? When does linking toxicity to the target actually matter? It is more likely that specific toxicities correlate with clinical outcomes when the target of the therapeutic agent is in fact a biologic driver of the disease. Renal cell carcinoma is characterized by loss of the von Hippel Lindau tumor suppressor gene activity, with resultant high expression of proangiogenic growth factors such as vascular endothelial growth factor.5 In a retrospective pooled analysis of four studies of an anti–vascular endothelial growth factor tyrosine kinase inhibitor (ie, sunitinib) for the treatment of metastatic renal cell carcinoma, an association was demonstrated for improved clinical outcomes in patients who developed sunitinib-associated hypertension.6 Likewise, the epidermal growth factor receptor is an integral part of the biology of KRAS wild-type metastatic colorectal carcinoma.7 In both a retrospective analysis and, more impressively, a prospective trial of dosing the anti–epidermal growth factor receptor monoclonal antibody panitumumab to skin toxicity, greater dermal toxicity was associated with improved survival.8,9
Toxicity may simply reflect the dose-related toxicity of the anticancer agent and a common pathway for both malignant and normal cells. Studies have not shown a linkage of efficacy with hematologic toxicity beyond a certain threshold. In a prospective trial of dosing adjuvant chemotherapy to neutropenia in early-stage breast cancer, no association was demonstrated with using severe neutropenia (grades 3 to 4) as a pharmacodynamic marker.10 Taxanes inhibit tubulin subunits with subsequent microtubule dysfunction, and there is not good evidence that microtubules are a dominant driver in the biology of breast cancer. Thus, treatment-related neuropathy would be expected to be a toxicity of the taxane but not necessarily a surrogate for efficacy.
Although clinical benefit may not be linked to neuropathy, a greater understanding of who may be at increased risk of developing severe toxicity is important. A number of researchers, including Schneider et al,1 have reported specific single nucleotide polymorphisms that seem to be associated with taxane toxicity and may account for the significant variation in neuropathy seen within our patient population.11–13 The ability to determine which individuals with a specific phenotype may be more susceptible to a specific toxicity may be of great relevance in terms of avoiding these severe adverse effects, especially if other treatment options exist.
In conclusion, although the development of taxane-induced neuropathy does not seem to be associated with improved clinical outcome, the potential use of single nucleotide polymorphisms to prospectively identify who develops neuropathy will be an important step forward in true individualized therapy. As the treatment of cancers moves toward a greater use of targeted agents, the identification of objective and standardized target toxicity as both pharmacokinetic and pharmacodynamic markers will be important in both selection of a sensitive population and customization of dose to efficacy.
AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
The author(s) indicated no potential conflicts of interest.
AUTHOR CONTRIBUTIONS
Manuscript writing: All authors
Final approval of manuscript: All authors
