Targeting Non-Mutant EGFR: Genetic and Clinical Biomarkers May Help Validate Use of Inhibitors

Oncology Live®, June 2012, Volume 13, Issue 6

A recent study suggests patients may be selected for therapy based on the number of EGFR gene copies, and evaluated for clinical benefit based on the severity of the rash that often develops.

Maurie Markman, MD

Editor-in-Chief of OncologyLive

Senior vice president for Clinical Affairs and National Director for Medical Oncology Cancer Treatment Centers of America, Eastern Regional Medical Center

It is well recognized that patients with non-small cell lung cancer whose malignancies possess an activating epidermal growth factor receptor (EGFR) gene mutation have a reasonably high probability of achieving a favorable response when treated with a tyrosine kinase EGFR inhibitor. It also is known that a small subset of patients with wild-type EGFR mutation may experience clinical benefit when treatment with this class of antineoplastics is delivered.

Unfortunately, it remains uncertain how clinicians can optimally select non-small cell lung cancer patients with nonmutant EGFR who have a reasonable probability of achieving a favorable outcome when treated with such drugs. This issue is particularly problematic for individuals with squamous cell lung cancer where activating mutations in EGFR rarely, if ever, occur.

In an effort to find clinical or biomarker features that may help predict for the beneficial impact associated with treatment with a tyrosine kinase EGFR inhibitor in individuals with a squamous cell lung cancer, investigators in Korea retrospectively studied a group of 71 patients treated with one of these agents in the second-line or later setting where tumor specimens were available for subsequent laboratory evaluation.1

In this analysis, patients found to have a high copy number of EGFR by fluorescence in situ hybridization testing were noted to have a far greater opportunity to achieve an objective response to treatment (26.3% vs 2%; P = .005) compared with individuals with a low EGFR copy number. Further, the relative risk of disease progression was shown to be reduced in this patient population (hazard ratio [HR], 0.57; P = .057) in a multivariate analysis. Of interest, in this same evaluation, patients who experienced at least a grade 2 rash (a well-recognized complication of tyrosine kinase EGFR inhibitors) were also observed to have a more favorable time to disease progression (HR, 0.54; P = .042) compared with patients with no rash or only a grade 1 rash.1

Finally, when these two clinical factors were analyzed together in this retrospective series, patients with a high EGFR copy number or at least a grade 2 rash experienced a substantially higher objective response rate (21.4% vs 0%; P = .003) and a longer time to disease progression (median 3.9 vs 1.13 months; P = .0002) compared with patients with a lower EGFR copy number and less than a grade 2 rash.1

If confirmed by others, these data may provide quite helpful support for clinicians considering second-line management options for patients with metastatic squamous cell lung cancer. For example, a decision to employ this class of drugs, as opposed to second-line cytotoxic chemotherapy, may be made in such a patient if a high EGFR copy number is demonstrated in molecular testing. Similarly, patients experiencing a rash after the initial treatment cycle may be counseled that this development suggests (but certainly by itself does not prove) this biological effect of the antineoplastic agent may subsequently be translated into a favorable clinical outcome (shrinkage of tumor masses, delay in disease progression).

Patients without EGFR gene mutations may benefit from tyrosine kinase inhibitors targeting EGFR. A recent study suggests patients may be selected for therapy based on the number of EGFR gene copies, and evaluated for clinical benefit based on the severity of the rash that often develops.

The issue of rash as a predictive clinically important factor in helping to define the subsequent utility of a targeted antineoplastic agent is worthy of additional comment. Not only does this information suggest the drug has achieved a sufficient concentration within an individual to impact a potentially relevant biological target in that patient, but the data also support the conclusion that the successful management of these specific unwanted signs and symptoms of the agent is a rational strategy for the individual in question, rather than discontinuation of the therapeutic approach (unless unquestionably necessary due to the severity of the toxic events).

It has similarly been suggested in certain clinical settings that the development of significant hypertension in an individual patient following treatment with an antiangiogenic agent may increase the statistical probability that a favorable therapeutic effect from the drug will be observed.2 Conversely, in other circumstances, the early onset of a clinically relevant adverse event in an individual patient (eg, peripheral neuropathy) may serve as a useful biomarker by predicting that more serious toxicity will likely develop if the treatment plan is continued (eg, maintenance paclitaxel chemotherapy in ovarian cancer), and alternative strategies should be strongly considered.3

References

  1. Lee Y, Shim HS, Park MS, et al. High EGFR gene copy number and skin rash as predictive markers for EGFR tyrosine kinase inhibitors in patients with advanced squamous cell lung carcinoma [published online ahead of print January 23, 2012]. Clin Cancer Res. 2012; 18(6):1760-1768.
  2. Scartozzi M, Galizia E, Chiorrini S, et al. Arterial hypertension correlates with clinical outcome in colorectal cancer patients treated with first-line bevacizumab [published online ahead of print October 7, 2008]. Ann Oncol. 2009;20(2):227-230.
  3. Markman M, Liu PY, Wilczynski S, et al. Phase III randomized trial of 12 versus 3 months of maintenance paclitaxel in patients with advanced ovarian cancer after complete response to platinum and paclitaxel-based chemotherapy: a Southwest Oncology Group and Gynecologic Oncology Group trial. J Clin Oncol. 2003; 21(13):2460-2465.