Advances in Diagnosis and Treatment of Non-Small Cell Lung Cancer (NSCLC)
This review document is periodically updated and reposted as new information is published. Please comment below with your suggestions for inclusion in upcoming updates of this review. (More PulmCCM topic reviews)
The diagnosis and treatment of non-small cell lung cancer (NSCLC) is rapidly evolving, although meaningful improvements in outcomes for most patients still elude us. New kinase inhibitors erlotinib, gefitinib and crizotinib are ushering in a potential new era of molecularly targeted therapies. And just when you thought you had the old one down, they've released a new TNM staging system for us to memorize (check out the ACCP's new online lung cancer staging tool for help with this).
Regarding multicentric nodules: Invasive sampling and pathologic examination of multicentric nodules is recommended in order to differentiate synchronous primary tumors from satellite nodules. Morphologic review by a pathologist is, so far, just as accurate at more high-tech methods like gene profiling at identifying separate lesions at telling synchronous primaries from satellite metastases.
Biomarkers for Early Diagnosis. Work is underway to learn how to identify resectable, early-stage lung cancers, although none of the methods are ready for wide use:
- Using microarray gene-expression profiling, it's possible to diagnose or rule out early stage lung cancer using bronchoalveolar lavage samples with negative cytology (normal-appearing cells), with sensitivity/specificity in the 80% range.
- Exhaled-breath condensate has been analyzed for volatile compounds as biomarkers, with sensitivity/specificity in the 70% range for early-stage lung cancer. Blood and urine markers have been less accurate.
Intense work has explored prognostic factors and/or biomarkers (intensity of PET-positivity; lab values like albumin; histologic cell type). After all that, the TNM stage, age, and (in stage I-II) sex remain the only important prognostic factors today.
Endoscopic mediastinal sampling. Where expertise in endoscopic ultrasound exists, a combined approach using esophageal and endobronchial bronchoscopic ultrasound (EUS/EBUS) to mediastinal lymph node sampling can achieve sensitivity of up to 88% and specificity up to 100%. This method permits highly accurate staging that is minimally invasive. Although this combined modality is used rarely in the U.S., it has the potential to largely replace mediastinoscopy as the first step in staging (with mediastinoscopy reserved for confirming EUS/EBUS-negative nodes are truly negative, an approach that increases sensitivity to 94%). At least 6 stations should be sampled and found negative, by whichever method, before pronouncing someone to have N0 disease.
Update on Treatments in Non-Small Cell Lung Cancer (NSCLC)
VATS vs. thoracotomy for NSCLC resection. Due to lack of randomization and suspected selection biases (e.g., early-stage NSCLC overrepresented in the VATS arm of trials), it's unknown whether there are any differences in outcomes after resection with video-assisted vs. open thoracotomy. Authors' bias is that thoracotomy is preferable and that VATS should be reserved for "selected patients with early-stage cancer."
Controversies in management of mediastinal-lymph node-positive NSCLC. (Super-summary: These continue.) IASLC data mining has suggested that patients with one zone of positive N2 nodes (stage III) have post-resection survival equal to people with multiple zones of N1-positive disease (stage II). Therefore, surgery should be offered to those with single-zone N2 disease. (Lymph node zones are different from stations. This is a new, more nuanced taxonomy that hasn't necessarily caught on yet.)
Induction chemotherapy +/- radiotherapy followed by surgery for "appropriate candidates" remains standard -- but who is appropriate (based on tumor-specific and patient-specific factors) is essentially determined site-by-site, is probably highly dependent on local physician preferences and practices, and thus wide heterogeneity in management of N2 disease exists. Authors favor lobectomy wherever possible (over chemoradiotherapy alone), based on improved observational survival (albeit in selected/biased patient sub-populations).
Adjuvant chemotherapy. Give cisplatin-based chemo after surgery to those with stage IB and higher, if possible (8% mortality reduction in stage IB, 17% in II and III).
Neo-adjuvant / induction chemotherapy: Randomized trials in the 1980s and 1990s showed benefits to chemotherapy before surgery. Subsequent trials have not met enrollment targets, but have not shown a benefit of neo-adjuvant chemotherapy. The best available meta-analysis suggests pre-operative chemotherapy confers a 5% absolute survival improvement at 5 years.
Testing for EGFR and EML4-ALK mutations; Tyrosine Kinase Inhibitors for NSCLC
Testing for EGFR mutations is becoming standard practice prior to initiation of therapy for NSCLC. In randomized trials including people with non-resectable NSCLC positive for the EGFR mutation, the tyrosine kinase inhibitors erlotinib (Tarceva) and gefitinib (Iressa) were as good or better as standard platinum based chemotherapy in increasing progression-free survival (without a true survival advantage). Erlotinib is currently FDA-approved for locally advanced or metastatic disease after standard chemotherapy; it showed a benefit in "all comers" in trials, and may be given without EGFR testing. (Gefitinib had less-impressive results in one trial, prompting the FDA to issue an advisory suggesting it should only be given to people with progressive NSCLC who've already shown a response to gefitinib.)
In August 2011, the FDA approved crizotinib (Xalkori), a kinase inhibitor for patients with locally advanced or metastatic NSCLC that express the ALK fusion gene (EML4-ALK). Crizotinib shrank tumors by 30% in slightly more than half of patients who had failed multiple chemotherapeutic regimens in randomized trials. Although it did not prolong actual survival, people taking crizotinib had longer lifespans than matched controls with NSCLC and the ALK mutation, in retrospective comparisons.
With gratitude to: Goldstraw P et al. Seminar: Non-small cell lung cancer. Lancet 2011;378:1272-1740.