How Do I Treat…?

Practical perspectives on cancer treatment by thought leaders, explaining how they would approach the treatment of a patient in their area of expertise.

Friday, December 28, 2012

ONLINE FIRST: GEOFFREY OXNARD: How I Use Molecular Testing in Clinical Practice for Patients with Lung Cancer

BY Geoffrey R. Oxnard, MD

Lowe Center for Thoracic Oncology,

Dana-Farber Cancer Institute;

Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School


The management of non-small cell lung cancer (NSCLC) has undergone remarkable changes over the past decade, and has significantly increased in complexity. There was a time when all first-line chemotherapy regimens were felt to be equivalent, such that an oncologist’s focus was on managing side effects to minimize morbidity. Subsequently, as the importance of histologic subtype became evident, NSCLC management grew more focused on the selection and sequencing of chemotherapy regimens. Today, the complexity of management has increased further, with a new focus on tumor genotyping to guide treatment with a growing number of oral targeted therapies.


The importance of tumor genotyping in NSCLC management means that oncologists now need to consider a myriad of complex questions when evaluating their patients’ potential to benefit from drugs like erlotinib or crizotinib. Hopefully, when considered one by one, each of these questions can be simplified for clinical practice:

  • Which patients should receive genotyping? I genotype any lung cancer patient who is a never-smoker (i.e., less than 5 packs total lifetime), regardless of histology. I genotype the majority of patients with NSCLC, excluding definite cases of squamous cell carcinoma. In addition to genotyping patients with advanced disease, I also genotype patients with stage IB-III NSCLC receiving definitive treatment -- these patients are at high risk of recurrence and may not be able to wait for genotyping should recurrence develop. Lastly, genotyping of multiple lung nodules can be a valuable strategy for distinguishing synchronous primaries (deserving definitive therapy) as opposed to metastatic lung cancer.
  • When should you order genotyping? The earlier genotyping is pursued, the greater the chance of it having an impact on management. For a patient with advanced NSCLC but no symptoms from the cancer, I will often delay treatment initiation until genotyping results are back. If a patient is symptomatic and chemotherapy can’t be delayed, I will pursue genotyping during first-line chemotherapy in preparation for maintenance or second-line therapy. I find genotyping to be important for guiding second-line therapy: Patients with a targetable genotype do better receiving a kinase inhibitor, while those without do better receiving second-line chemotherapy.
  • Which genes should be tested? EGFR and ALK genotyping are now a routine part of NSCLC care, as these tests indicate which patients should receive an EGFR or ALK inhibitor. I find that KRAS genotyping is also valuable because KRAS-mutant cancers do not benefit from existing kinase inhibitors and do not need further genotyping, because KRAS mutations are non-overlapping with other targetable genotypes. Many practitioners now also test for BRAF mutations, ROS1 rearrangements, and MET amplification because highly active therapies (vemurafenib and crizotinib) are commercially available for these genotypes. Testing for alterations in other cancer genes (HER2, PIK3CA, RET, etc.) can facilitate clinical trial enrollment.
  • What genotyping assay should be used? There is a diversity of genotyping assays being used by different commercial laboratories; all are adequate, though each has its weaknesses. For example, many genotyping assays are mutation-specific, meaning they might test for the most common EGFR mutations but might overlook less common EGFR mutations. Some other assays are less sensitive, making them vulnerable to falsely negative results when a poor quality biopsy is tested. Still other assays are highly sensitive for detecting mutations in a subpopulation of tumor cells, but are vulnerable to false positive results. Some academic centers are beginning to use targeted next-generation sequencing (NGS) assays  which are very accurate and can simultaneously test for point mutations, insertions, rearrangements, and amplifications in a variety of different cancer genes; while NGS takes longer and costs more than EGFR and ALK genotyping alone, the comprehensive results make it a compelling approach.
  • What tissue is best to genotype? Archived paraffin-embedded tissue is acceptable because these targetable genotypes in NSCLC generally persist over the entirety of a cancer’s course, and do not change with time. Surgical specimens are reliable unless there is significant necrosis or fibrosis, such as after radiation. Core biopsies are generally adequate except for bone biopsies, because the decalcification process leads to poorer DNA quality. Malignant pleural effusions can be ideal because of their easy availability; I routinely ask that these be spun into cell blocks and archived for future use if biopsy tissue is scarce. Fine needle aspirates are more variable in their adequacy but definitely are worth testing if no other biopsy tissue is available.
  • When should a biopsy be ordered? Any NSCLC patient whose diagnostic biopsy specimen is inadequate for standard EGFR and ALK genotyping needs a new biopsy to determine the role of targeted therapies in their care. In patients who require immediate treatment, chemotherapy can be started and the biopsy should be scheduled between chemotherapy cycles; I recommend against waiting until progression to perform a biopsy because patients may be too unwell at that point in their care. In patients with “wild-type” NSCLC who have completed second-line chemotherapy and are still interested in further therapy, a biopsy for identification of a targetable genotype may be the intervention most likely to prolong their life given the minimal activity of erlotinib in EGFR wild-type cancers and the limited data on other third-line therapies.
  • What about a repeat biopsy after progression on targeted therapies?  Resistance to erlotinib and crizotinib in lung cancers with EGFR and ALK alterations is an active area of investigation, though targeted treatments for resistance have not made it into clinical practice. Currently, the most compelling reason to biopsy after development of resistance is that a subset of patients have been found to exhibit transformation to a small cell histology, and therefore may be best treated with platinum and etoposide. Genotyping of resistance biopsies may also demonstrate new resistance mutations; the most common resistance mechanism to erlotinib, the EGFR T790M mutation, has been shown to be associated with slower growth and a relatively favorable prognosis after resistance develops. Perhaps most important, resistance biopsies can help to guide clinical trial selection for this growing population of lung cancer patients.