Azzoli, Christopher G. MD*; Park, Bernard J. MD†; Pao, William MD, PhD*; Zakowski, Maureen MD‡; Kris, Mark G. MD*
Nonsmall cell lung cancer (NSCLC) is the major cancer killer worldwide, accounting for more than 1.2 million deaths each year. NSCLC is deadly because it is usually a systemic disease at the time of presentation. Even after complete surgical resection of stage I–III NSCLC, approximately half of patients will recur and die within 5 years because of unrecognized micrometastatic disease.1 Recent clinical trial data demonstrate that 4 months of cisplatin-based adjuvant chemotherapy improves survival in patients with stage IB–III NSCLC after complete resection.2–4
The data in favor of adjuvant chemotherapy for patients with resected NSCLC have limitations. The clinical trials randomized fewer than 5000 patients with a median follow-up of 5 years. In comparison, adjuvant therapy for breast cancer is guided by data from more than 110,000 patients with more than 15 years of follow-up.5 Historically, the study of adjuvant chemotherapy in NSCLC has been hampered by nihilism from past negative trials, difficulties delivering chemotherapy to patients after thoracotomy, and the fact that lung cancer patients are older (median age 70 in the United States), and have higher rates of comorbid medical illnesses related to cigarette smoking than other cancer patients.
The patients enrolled in the positive clinical trials were mostly younger (median age 60), had high-risk NSCLC (stage IB–III), good performance status, and could safely receive cisplatin.2–4 In the real world, oncologists see NSCLC patients who are generally older, with clinical or pathologic characteristics which may have excluded enrollment on the pivotal trials. For example, there is currently no data to support the use of adjuvant chemotherapy in patients with stage IA NSCLC, little support for treating IB patients, no data for stage II–III patients who are not candidates for cisplatin, or stage IV patients rendered disease-free by surgery. There is no data to support chemotherapy in patients with a positive surgical margin, patients with pure bronchioloalveolar subtype, or for patients aged more than 75 years.6 All patients in previous randomized trials began their chemotherapy within 2 months of their date of surgery, so there is no data to support starting chemotherapy more than 2 months postoperatively. Outside of studies of oral fluoropyrimidines conducted in Japan, there are no data to support noncisplatin-based adjuvant chemotherapy.7,8
For patients who do not fit the profile of those enrolled on the prior clinical trials, there are no data supporting adjuvant chemotherapy. One might consider treating borderline patients (stage I, stage IV, or any patient aged >75 years) except for the substantial risks of adjuvant chemotherapy. Cisplatin-based adjuvant chemotherapy causes side effects including venous irritation from vinorelbine (20–30%), debilitating fatigue (15–28%), nausea (10–27%), anorexia (10–15%), anemia (7–14%), infection (1–11%), febrile neutropenia (7–9%), hair loss (0–5%), peripheral neuropathy (3%), and a risk of treatment-related death (1–2%).2,4 As a result, half of the patients drop out before completing all 4 months of chemotherapy.
What should the oncologist recommend for a young, fit patient with a 2.8 cm tumor (borderline IB), or an elderly patient with resected stage III NSCLC who is unlikely to tolerate cisplatin? Identification of biomarkers beyond pathologic stage could help inform the oncologist as to which patients to treat, and which agents to use.
CLINICAL PROGNOSTIC AND PREDICTIVE FACTORS
Prognostic factors predict outcome (usually survival) independent of the treatment administered, and can be used to classify patients as high-risk, or low-risk. Predictive factors predict response of the tumor to treatment, usually in terms of tumor shrinkage, or survival benefit from treatment.
To date, pathologic stage is the only prospectively validated prognostic and predictive clinical factor for prescribing adjuvant chemotherapy for NSCLC. A meta-analysis of data from more than 5000 patients randomized to cisplatin-based adjuvant chemotherapy versus surgery alone (Lung Adjuvant Cisplatin Evaluation), demonstrated that higher stage patients garner more benefit from adjuvant chemotherapy.9 Patients with stage II–III disease clearly benefit from chemotherapy (hazard ratio [HR], 0.83; 95% confidence interval [CI], 0.73–0.95). There is less benefit seen in patients with stage IB disease (HR, 0.92; 95% CI, 0.78–1.10), and potential for harm in patients with stage IA disease (HR, 1.41; 95% CI, 0.96–2.09).
Therefore, stage is the most important factor to consider when prescribing adjuvant chemotherapy. The higher the stage, the worse the prognosis, and the more likely adjuvant chemotherapy may be beneficial (Figure 1). Other prospectively validated clinical factors which suggest better prognosis include young age, squamous histology in early-stage patients (IB–II), and whether the patient received adjuvant chemotherapy.2–4 Other than stage, there are no clinical factors which predict which patients will benefit more from adjuvant chemotherapy (Table 1).
MOLECULAR PROGNOSTIC AND PREDICTIVE FACTORS
Patients who have undergone resection of NSCLC usually have adequate tissue for molecular analysis. Rapid advances in technology have lead to high-throughput assays to measure changes in NSCLC DNA, RNA, and proteins so as to identify potential molecular biomarkers of clinical outcome. The discovery of molecular markers of potential significance, and refinements in molecular assays, will likely outstrip our ability to test all of them in prospective clinical trials. Almost all current molecular markers of interest have been developed using tissue banks, and retrospective cohort studies (Table 2).
Kirsten Rat Sarcoma Viral Oncogene Homolog (KRAS)
The only molecular marker which has been studied prospectively in a randomized trial of adjuvant chemotherapy is mutation in rat sarcoma viral oncogene (RAS). RAS is an enzyme (GTPase) which transmits growth signal by binding to, and hydrolyzing guanosine triphosphate. RAS is downstream of the growth signal transmitted into a cell by the epidermal growth factor receptor (EGFR). Three human RAS genes have been identified—HRAS (Harvey), KRAS (Kirsten), and NRAS (neuroblastoma)—each of which is highly homologous. RAS mutations are detectable in approximately 20% of lung cancers, are found more often in cigarette smokers, and have been associated with a poor prognosis in several studies.10–15 KRAS mutations constitute 90% of all RAS mutations in NSCLC. More than 80% of KRAS mutations occur in codon 12, whereas other mutations are located in codons 13 and 61.
In the National Cancer Institute of Canada JBR-10 adjuvant trial, 482 patients with completely resected stage IB to IIB NSCLC were analyzed for RAS mutations.2 Participating centers submitted fresh-frozen primary tumor or paraffin-embedded blocks of tissue specimens to a central laboratory for RAS mutation analysis of codons 12, 13, and 61 of the HRAS, KRAS, and NRAS genes by allele-specific oligonucleotide hybridization. The RAS mutations found were confirmed by direct sequencing. The status of RAS mutations in the tumors was able to be determined in 450 patients (93%), with 24% of these patients found to have a RAS mutation in their tumor. Patients were prospectively stratified by the presence of RAS mutation, allowing retrospective subgroup analysis.
The median survival among patients with wildtype RAS in the observation group was 74 months and was not reached in the group that received chemotherapy (HR, 0.69; 95% CI 0.49–0.98; p = 0.03). However, for patients whose tumors harbored a RAS mutation, adjuvant chemotherapy did not confer a survival advantage (HR, 0.95; 95% CI 0.53–1.71; p = 0.87). In an interaction analysis, the effect of the status of RAS mutations on the outcome of treatment was not statistically significant (p = 0.29), and RAS was not an independent predictor of survival in this study. However, the lack of demonstrable benefit of adjuvant cisplatin plus vinorelbine chemotherapy in RAS-positive patients in this study, despite the large benefit seen overall in this trial, suggests that patients with a RAS mutation should not be offered postoperative cisplatin plus vinorelbine chemotherapy.
EGFR—also known as human epidermal growth factor receptor 1 (HER1), or avian erythroblastic leukemia viral oncogene homolog (ErbB1)—is a signaling protein attached to the cell membrane that transmits growth signals into the cytoplasm through a tyrosine kinase which is activated when the receptor binds to growth factors. The anticancer drugs, erlotinib and gefitinib, are small molecule inhibitors of the EGFR tyrosine kinase. Erlotinib is currently approved as a second-, or third-line chemotherapy for patients with metastatic NSCLC based on a prospective trial showing an improvement in overall survival versus placebo.16 In 10% of patients with radiologic response to erlotinib, the median duration of response was 8 months. For all patients treated with erlotinib, median overall survival was 7 months compared with 5 months with placebo (HR, 0.70; p < 0.001).
In 2004, three groups of investigators independently discovered that mutations in the tyrosine kinase domain of EGFR are strongly associated with sensitivity to erlotinib and gefitinib.17–20 These mutations are almost never found in NSCLC with squamous histology, but occur in approximately 10% of adenocarcinomas. Patients with EGFR mutant tumors live longer than patients with EGFR wildtype tumors, regardless of the treatment. In one retrospective study, patients with metastatic NSCLC and no EGFR mutation had a median survival of 10 months, while the median survival of patients with EGFR mutation was not reached.21 The superior outcome of patients with NSCLC and EGFR mutation in this study confounded Kaplan-Meier analysis of survival given that so few deaths were observed in these patients.
The different types of EGFR mutations have different prognostic values. Patients with EGFR mutations characterized by deletion in exon 19 have a better prognosis than patients with missense mutations in EGFR exon 21.22,23 In one retrospective study of 34 patients with EGFR mutation treated with erlotinib or gefitinib, patients with EGFR exon 19 deletions had significantly longer median survival than patients with EGFR L858R mutations (34 versus 8 months; log-rank p = 0.01).23
There are numerous, prospective, single-arm clinical trials which suggest that erlotinib or gefitinib are highly active treatments for patients whose lung tumors harbor an EGFR mutation.24–27 Whereas it is well-established that the presence of an EGFR mutation may increase responsiveness of NSCLC to erlotinib or gefitinib, these same mutations predict longer survival, regardless of therapy, which may confound detection of the effect of treatment on overall survival. As such, there is ongoing debate about which EGFR biomarker—mutation, gene amplification, or protein overexpression—is more important to determine prognosis, and to predict benefit from treatment with erlotinib and gefitinib.28,29
To date, there are no prospective, randomized studies to prove that erlotinib or gefitinib improves survival in patients with metastatic NSCLC and an EGFR mutation. A retrospective study of patients enrolled in a placebo-controlled trial of erlotinib found that EGFR gene copy number was a better predictor of survival benefit from erlotinib than EGFR mutation.28 These data have been criticized due to the potential high false-positive rate of EGFR mutations, and small subgroup of patients studied (197 tissue samples from 731 patients on the trial), which may have confounded this analysis.
Subgroup analyses from several trials of gefitinib and erlotinib in patients with metastatic NSCLC have confirmed that patients with NSCLC who have never smoked cigarettes benefit from these drugs more than patients who smoke,16,30,31 and never smokers are much more likely to have EGFR mutations than former or current smokers.17,32 As a result of this debate, EGFR mutation analysis, EGFR overexpression by immunohistochemistry (IHC), EGFR amplification analysis by fluorescent, or chromogenic in situ hybridization (FISH/CISH), and never-smoking status are all currently being studied in prospective trials of patients with metastatic NSCLC receiving erlotinib and gefitinib. Evolving data suggest that tumors with EGFR amplification are more likely to harbor EGFR mutations, and vice versa, confounding the comparison between these two molecular characteristics.
Of note, mutations in KRAS and EGFR are mutually exclusive. Furthermore, the presence of a KRAS mutation predicts a lack of benefit from gefitinib and erlotinib.21,33,34 As such, patients whose NSCLC harbors a known KRAS mutation need not be screened for an EGFR mutation, and vice versa.
A retrospective analysis of 296 patients who underwent resection at the Memorial Sloan-Kettering Cancer Center for stage I–III lung adenocarcinoma using established techniques identified EGFR and KRAS mutations in tumors from 40 (13%) and 50 (17%) of the patients, respectively. (J. Marks and W. Pao, personal communication.) No tumor had both mutations. EGFR mutations were significantly associated with a history of never smoking, whereas KRAS mutations were associated with smoking (p < 0.005). None of the patients received gefitinib or erlotinib. After adjustment for pathologic stage, patients with EGFR mutations displayed a trend toward longer survival when compared with patients with KRAS mutations (p = 0.11). These data suggest that EGFR and KRAS mutations define clinically distinct molecular subsets of lung adenocarcinoma.
The RADIANT study (Randomized Double-Blind Trial In Adjuvant NSCLC with Tarceva) is an ongoing, multicenter, placebo-controlled phase 3 clinical trial of adjuvant erlotinib in patients with resected NSCLC which expresses EGFR by IHC, or demonstrates EGFR amplification by FISH.35–37 Tissue will be studied for EGFR and KRAS mutations retrospectively. By allowing enrollment of either IHC or FISH-positive patients, and checking EGFR and KRAS mutation status retrospectively, this trial design has been criticized as failing to enrich the study population with patients most likely to benefit from erlotinib. As RADIANT continues, there is a need for smaller, single-arm phase 2 clinical trials of erlotinib or gefitinib as adjuvant chemotherapy, using alternative entrance criteria, such as EGFR mutations. Much may be learned by comparing the results of these smaller studies to the treatment arm of RADIANT.
Excision Repair Cross-Complementation Group 1 (ERCC1)
More than 800 resected NSCLC tumors from patients enrolled in the International Adjuvant Lung Trial (IALT) have been analyzed retrospectively for the expression of 19 independent genes by routine IHC.3,38 Categories of the 19 molecular markers include DNA repair genes, drug transporters, signal transduction genes, apoptosis, and cell cycle regulators. This work is ongoing, and much of the data have not been reported.
Cisplatin kills cancer cells by binding covalently to DNA to form adducts, which interfere with normal DNA replication. This DNA damage may trigger apoptosis in dividing cells. Nucleotide excision repair is the major pathway for repairing DNA damaged by cisplatin. ERCC1 is the rate-limiting enzyme in this repair process. Cancer cells which overexpress ERCC1 are more likely to have de novo resistance to cisplatin.
Among 761 tumors assayed in the IALT for ERCC1 expression by immunohistochemistry, 335 (44%) were scored as positive, and 426 (56%) were scored as negative.39 The absence of ERCC1 was associated with increased benefit from cisplatin-based adjuvant chemotherapy (test for interaction, p = 0.009). Compared with surgery alone, adjuvant chemotherapy significantly prolonged survival among patients with ERCC1-negative tumors (HR, 0.65; 95% CI 0.50–0.86; p = 0.002) but not among patients with ERCC1-positive tumors (HR, 1.14; 95% CI 0.84–1.55; p = 0.40).
Among patients who did not receive adjuvant chemotherapy, those with ERCC1-positive tumors survived longer than those with ERCC1-negative tumors (HR, 0.66; 95% CI 0.49–0.90; p = 0.009). A multivariate logistic model showed that the expression of ERCC1 was significantly correlated with age (p = 0.03; less common in patients <55 years of age than in patients 55–64 years of age), histologic type (p < 0.001; less common in adenocarcinomas than in squamous-cell carcinomas), and pleural invasion (p = 0.01; less common in the absence than in the presence of pleural invasion).39
An independent study used an automated, quantitative method to measure expression of ERCC1—and 2 other genes, ribonucleotide reductase subunit 1 (RRM1), and phosphatase and tensin homolog—in 187 patients who had received only surgical treatment for NSCLC.40 The investigators confirmed that RRM1 expression correlated with the expression of ERCC1 (p < 0.001). There is retrospective evidence that low RRM1 expression is a poor prognostic factor in NSCLC.41 In addition, RRM1 is the predominant cellular determinant of the efficacy of the nucleoside analogue gemcitabine (2′,2′-difluorodeoxycytidine), an important chemotherapy drug used for the treatment of NSCLC which interferes with ribonucleotide reductase.42,43 In the study by Zheng et al, the overall survival was more than 120 months for patients with tumors with high expression of RRM1, and 60 months for those with low expression of RRM1 (HR, 0.61; p = 0.02). Among these 187 patients, the survival advantage was limited to the 30% of patients with tumors that had a high expression of both RRM1 and ERCC1.
The clinical relevance of ERCC1 and RRM1 expression has also been studied extensively in patients with metastatic NSCLC. In patients with metastatic NSCLC, ERCC1 expression predicts a lower radiologic response rate to platinum-based chemotherapy, and worse survival.44,45 Prospective clinical trials have been conducted which assign cisplatin, versus noncisplatin chemotherapy based on ERCC1 expression. One study randomized 444 patients with stage IV NSCLC to blind treatment, versus treatment assigned based on ERCC1 expression. RNA was isolated from pretreatment biopsies, and quantitative reverse-transcriptase polymerase chain reaction was performed to determine ERCC1 mRNA expression. Patients were randomly assigned in a 1:2 ratio to either the control or genotypic arm before ERCC1 assessment. Patients in the control arm received docetaxel plus cisplatin. In the genotypic arm, patients with low ERCC1 levels received docetaxel plus cisplatin, and those with high levels received a nonplatinum regimen (docetaxel plus gemcitabine). Of 346 patients assessable for response, objective response was attained by 53 patients (39%) in the control arm and 107 patients (51%) in the genotypic arm (P = 0.02).46
Another study tested tumor tissue for both ERCC1 and RRM1, and assigned patients to four different chemotherapy regimens (with or without cisplatin, and with or without gemcitabine). Sixty of the 85 patients enrolled underwent biopsy and gene expression analysis. The overall rate of major radiologic response was 44%, survival proportion at 1 year was 59%, and median survival time was 13 months, suggesting that therapeutic decision-making based on RRM1 and ERCC1 gene expression for patients with advanced NSCLC is feasible and promising for improvement in patient outcome.47
Given the results in the adjuvant setting, and early results assigning chemotherapy to patients with metastatic NSCLC based on ERCC1 status, it is reasonable to conduct clinical trials which offer noncisplatin chemotherapy to patients with ERCC1-positive NSCLC. There are plans for a phase 2 (feasibility) study by the Southwest Oncology Group to treat patients with adjuvant chemotherapy based on ERCC1 and RRM1 expressions. Patients with tumors with low RRM1 and low ERCC1 will receive gemcitabine plus carboplatin. Patients with low RRM1 and high ERCC1 will receive gemcitabine with docetaxel (a noncisplatin combination). Patients with high RRM1 and high ERCC1 will receive vinorelbine and docetaxel (a nongemcitabine, noncisplatin combination). Patients with high RRM1 and low ERCC1 will receive docetaxel and carboplatin.
The cell cycle regulators measured in IALT tumors included p27Kip1, p16, cyclin D1, cyclin D3, cyclin E, and Ki-67.48 Among these six proteins, only p27 staining had predictive value. p27Kip1—also known as Cdkn1b (cyclin-dependent kinase inhibitor 1B)—is a cyclin-dependent kinase inhibitor which has an antiproliferative effect on cells, and which may protect cells from apoptosis by fostering growth arrest. Thus, p27 overexpression may impart de novo resistance to cisplatin by allowing the cancer cell to repair cisplatin-induced DNA damage. In patients with p27-negative tumors, cisplatin-based chemotherapy resulted in longer overall survival compared to controls (HR, 0.66; 95% CI 0.50–0.88; p = 0.006). In patients with p27-positive tumors, overall survival was not different between patients treated with cisplatin-based chemotherapy and controls (HR, 1.09; 95% CI 0.82–1.45; p = 0.54). None of these cell cycle genes were significantly associated with overall survival in the total study population. However, historical data consistently demonstrate that p27-positivity is associated with improved prognosis.49–52 Thus, both ERCC1 and p27 expressions have parallel prognostic and predictive trends—positive staining means a better prognosis, and less benefit from cisplatin-based chemotherapy.
When ERCC1 and p27 staining results are combined, the predictive values are further increased.38 Patients negative for both proteins benefit the most from adjuvant chemotherapy (HR, 0.52, 95% CI 0.36–0.74), whereas patients positive for both seem more likely to be harmed than helped by adjuvant chemotherapy (HR, 1.27, 95% CI 0.87–1.84). However, the staining for ERCC1 and p27 fails to correlate in nearly half of the patients (42%)—i.e., very often tumors are positive for one gene, but negative for the other.
Other Potential Biomarkers
Tubulins are a family of globular proteins that make up microtubules in cells, vital for cell structure, movement, mitosis, and metabolism (vesicular transport). High expression of class III beta-tubulin (bTubIII) in advanced NSCLC is known to correlate with reduced response rates and inferior survival with antimicrotubule agents. Tumor tissue from 265 of the 482 patients enrolled in the NCIC-JBR.10 trial was analyzed for bTubIII expression by IHC.53 High bTubIII expression was associated with poorer survival in patients treated with surgery alone, but not in patients treated with adjuvant chemotherapy. The benefits of adjuvant chemotherapy were greater in high versus low tubulin expressors, however this interaction did not reach statistical significance.
Multidrug resistance proteins are membrane proteins which pump foreign molecules (including drugs) out of cells. Two multidrug resistance proteins (MRP1, MRP2) have been tested in tissue from IALT.54 MRP2-positive patients (47%) had a significantly worse prognosis, making it an independent prognostic factor (HR, 1.37, 95% CI 1.09–1.72, p = 0.007). However, neither MRP1 nor MRP2 had predictive value with regard to benefit from adjuvant chemotherapy.
Apoptotic markers tested in tissue from IALT include the death receptor Fas, its ligand FasL, and survivin.55 Fas (tumor necrosis factor receptor superfamily, member 6), and the ligand for Fas (FasL), are triggers for both intrinsic and extrinsic programmed cell death (apoptosis). Survivin is an inhibitor of apoptosis which regulates chromosome segregation and cytokinesis during mitosis. Of 773 evaluable cases of NSCLC from IALT, 73% were Fas negative, 49% FasL negative, and 46% survivin negative. Fas, FasL, and survivin scores were not related to prognosis, but the ratio of a Fas to FasL ≥1 was related to longer survival (HR = 0.72; p = 0.02). There was a trend which suggested that chemotherapy was more likely to benefit FasL-negative patients (HR = 0.69) as compared to FasL-positive (HR = 1.03; p = 0.06), as well as for Fas:FasL ratio >1 (HR = 0.51) as compared to a ratio of ≤1 (HR = 0.80; p = 0.05).
Analyses of other biomarkers in tissue from IALT is ongoing, including mutations of p53, KRAS, and EGFR.38 It is hoped that all of the results from IALT, including the promising results measuring ERCC1 and p27, will be validated in tissue from other adjuvant trial populations, such as the Adjuvant Navelbine International Trialist Association trial, and NCIC-JBR-10. Until these biomarkers have been validated in independent studies, they should not be used as the sole criteria to deny adjuvant chemotherapy to patients with completely-resected stage II–III NSCLC who are eligible for cisplatin. All of the genes mentioned above are certainly worthy of study in clinical trials, and there may already be a role for KRAS or ERCC1 to tip the scales in patients where there is controversy (i.e., a younger patient with stage I NSCLC, or any patient aged >75 years).
Gene Sets and Expression Arrays
Numerous studies have compared global gene expression profiles from high-risk and low-risk NSCLC tumors to identify sets of genes whose coordinate expression pattern may predict patient outcome.56–65 Both the validity and reproducibility of microarray-based clinical research have been challenged due to the potential for spurious results from multiple testing, or incorrect cross-validation procedures.66 Other ongoing debates include whether overall survival, disease-free survival, or disease-specific survival are the best endpoints to use in these analyses, and whether to separate different histologies (adenocarcinoma from squamous carcinoma).
There is a disconcerting lack of overlap between the genes identified in the different experiments, with few or no genes in common in independent gene signatures. For example, of the 260 genes identified in five published gene signatures, only three genes were common to more than one signature.57 This lack of overlap is also seen in gene expression signatures for predicting outcome of breast cancer patients.67 This suggests that several different gene expression signatures are capable of predicting outcome, and raises the question of which platform is the most accurate.
Clarity may come from combining datasets to get larger sample sizes, identifying the biologic processes behind the different gene expression patterns, or correlating gene expression profiles with other powerful prognostic or predictive markers such as EGFR or KRAS mutation, or ERCC1 expression. The technique for handling tissue and the platform for measuring gene expression is vital to the results achieved.68,69 Moving forward, it may be more important to demonstrate reproducible results using the same platform, including standardized processing of tissue and RNA expression measurement, than to reconcile the disparate gene sets derived from disparate platforms.
Given the number of drugs available for the treatment of NSCLC, gene expression signatures may ultimately prove to be more valuable if they carry both prognostic and predictive information. None of the studies of NSCLC gene signatures published to date inform choice of chemotherapy. Early experiments using lung cancer cell lines with variable sensitivity to different drugs demonstrate the ability to identify oncogenic pathways which predict which drugs will be more beneficial for treatment.70 This approach will require validation in patients with metastatic disease, in whom drug efficacy is immediately apparent, before it can be applied in the adjuvant setting.
Pilot Program to Study Molecularly Tailored Adjuvant Chemotherapy
Given the potential of EGFR mutations, KRAS mutations, and ERCC1 expression in informing adjuvant therapy, the multidisciplinary Thoracic Disease Management Team at MSKCC, has developed standardized methods for measuring these molecular markers in all resected NSCLC specimens (Table 3). We have established sensitive, nonsequencing-based polymerase chain reaction assays to detect the common mutations in EGFR exons 19 and 21, which account for 90% of EGFR mutations found in NSCLC.71 We are currently testing all resected lung adenocarcinomas for EGFR mutations in exons 19 and 21 by this polymerase chain reaction-based assay,71 and KRAS mutation by direct sequencing of exon 2. We are only testing adenocarcinomas for these mutations given the low rates of detection of these mutations in squamous tumors. Genomic DNA for testing can be extracted from 10 to 15 unstained tumor-bearing slides, paraffin-embedded tissue blocks, or freshly frozen tumor specimens. We are also developing mass-spectrometry-based methods (Sequenom, Inc., SanDiego, CA) which may prove to be a more efficient method for identifying point mutations in tumor DNA.72 Measurement of ERCC1 protein expression is being performed by IHC on paraffin-embedded NSCLC tissue using the methods of Olaussen, et al.39
The value of these biomarkers is that they are both prognostic, to help decide which patients should be offered adjuvant therapy, and predictive, potentially important in deciding which treatment to assign. Patients are offered enrollment in clinical trials of adjuvant therapies based on their molecular phenotype (Table 4).
Based on available data, cisplatin plus vinorelbine is the best, and most studied drug regimen for adjuvant treatment of NSCLC. Randomized trials to establish alternative or superior chemotherapy will require large numbers of patients, and years of follow-up. Meanwhile, the list of promising drugs for the treatment of metastatic NSCLC continues to grow. There is strong retrospective data to suggest that patients with high ERCC1 expression in their resected NSCLC should not be offered cisplatin-based adjuvant chemotherapy, and validation of the efficacy of noncisplatin regimens is sorely needed.
In collaboration with investigators at the Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins, MSKCC is conducting a phase 2 study of a noncisplatin drug combination (vinorelbine plus docetaxel) as adjuvant therapy for patients with NSCLC which expresses ERCC1. This regimen was chosen based on phase 2 data in patients with metastatic NSCLC showing efficacy comparable to other noncisplatin combinations, and toxicities—namely onycholysis and hyperlacrimation—which accompany prolonged administration and may be avoided in a fixed course of adjuvant chemotherapy.73,74 Neutropenia from this regimen is manageable with growth factor support.
Similarly, there is a strong argument not to offer standard adjuvant therapy to patients with NSCLC which harbors a KRAS mutation, but instead explore RAS-directed therapies. At MSKCC we are conducting a phase 2 study to document the immunogenicity of GI-4000 (GlobeImmune, Inc., Louisville, CO), a recombinant, heat-inactivated yeast (Saccharomyces cerevisiae) engineered to express one of three mutated RAS oncoproteins.75 In combination with genomic sequencing of tumors before initiation of immunotherapy, this approach has potential for immediate application to human cancers driven by mutations in KRAS.75
Finally, patients with EGFR exon 19 or 21 mutations in their resected adenocarcinoma will be offered enrollment in a clinical trial of erlotinib as adjuvant therapy, based on the prospective phase 2 data in metastatic NSCLC documenting very high radiologic response rates in this molecular subgroup. We plan to compare the efficacy results of our single-arm trial to the outcomes of patients enrolled in the placebo-controlled RADIANT trial.
Given the pace at which new drugs are being discovered for the treatment of metastatic NSCLC, and the speed at which faster, cheaper, and more accurate techniques for molecular analysis of tumor tissue are being developed, it is likely that a promising molecular test designed to guide a specific treatment may become obsolete by the time the approach has been validated in a prospective trial. This is especially true in the adjuvant setting, where prospective clinical trials require many years to complete. Progress in this field will require clinical judgment, and rational extrapolation of retrospective data, including data from patients with metastatic NSCLC, given that it would be impossible to anticipate or test every new technology, and every clinical situation in a prospective adjuvant trial.
The current retrospective data are not sufficient to support the routine use of molecular markers to guide adjuvant therapy for NSCLC outside of a clinical trial. We await validation of ERCC1, and p27 in tissue from Adjuvant Navelbine International Trialist Association and JBR10, and validation of KRAS and EGFR mutation analysis in IALT. We also look forward to validation of the usefulness of adjuvant erlotinib, now being tested in the RADIANT study.
Pathologic stage remains the most important factor to consider when prescribing adjuvant chemotherapy. The presence or absence of a molecular characteristic should not be the sole criterion to withhold adjuvant cisplatin plus vinorelbine in patients with completely resected stage II to III NSCLC who are fit and otherwise eligible to take this chemotherapy safely. However, for patients in whom the risk:benefit ratio is less clear—such as patients with stage I NSCLC, or patients with stage II to III disease who are aged more than 75 years and where cisplatin may be more dangerous—EGFR mutation testing, KRAS mutation testing, or ERCC1 expression by IHC, may tip the scales.
Before a molecular test can be adopted for routine practice, a valid and standardized laboratory technique must be established. For example, there is recent debate with regard to the specificity of the monoclonal mouse antibody (mAb 8F1) which was used to detect ERCC1 in the tumor tissue from IALT,76 and there is a lack of association between ERCC1 expression as measured by messenger RNA levels by reverse-transcriptase polymerase chain reaction, versus protein expression by IHC.40 Also, there are differing opinions regarding the most efficient and accurate method for detecting EGFR mutations in tumor DNA.77
Pilot studies of molecularly tailored adjuvant therapy for patients with resected NSCLC being conducted at MSKCC and elsewhere will help to validate laboratory techniques for molecular assays, establish the feasibility of this approach, and serve as a stimulus for others to explore additional innovations in our management of persons with resected NSCLC who remain at risk for relapse. These phase 2 data will inform the design of future phase 3 trials in which novel therapies are assigned on the basis of genetic tests, with the goal to improve survival for patients with both early-stage and metastatic NSCLC.
1. Ravdin PM, Davis G. Prognosis of patients with resected non-small cell lung cancer: impact of clinical and pathologic variables. Lung Cancer
2. Winton T, Livingston R, Johnson D, et al. Vinorelbine plus cisplatin vs. observation in resected non-small-cell lung cancer. N Engl J Med
3. Arriagada R, Bergman B, Dunant A, Le Chevalier T, Pignon JP, Vansteenkiste J. Cisplatin-based adjuvant chemotherapy in patients with completely resected non-small-cell lung cancer. N Engl J Med
4. Douillard JY, Rosell R, De Lena M, et al. Adjuvant vinorelbine plus cisplatin versus observation in patients with completely resected stage IB-IIIA non-small-cell lung cancer (Adjuvant Navelbine International Trialist Association [ANITA]): a randomised controlled trial. Lancet Oncol
5. Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomised trials. Lancet
6. Pepe C, Hasan B, Winton TL, et al. Adjuvant vinorelbine and cisplatin in elderly patients: National Cancer Institute of Canada and Intergroup Study JBR. 10. J Clin Oncol
7. Kato H, Ichinose Y, Ohta M, et al. A randomized trial of adjuvant chemotherapy with uracil-tegafur for adenocarcinoma of the lung. N Engl J Med
8. Strauss GM, Herndon JE, Maddaus MA, et al. Adjuvant chemotherapy in stage IB non-small cell lung cancer (NSCLC): Update of Cancer and Leukemia Group B (CALGB) protocol 9633. J Clin Oncol
9. Pignon JP, Tribodet H, Scagliotti GV, et al. Lung Adjuvant Cisplatin Evaluation (LACE): a pooled analysis of five randomized clinical trials including 4584 patients. J Clin Oncol
10. Ahrendt SA, Decker PA, Alawi EA, et al. Cigarette smoking is strongly associated with mutation of the K-ras gene in patients with primary adenocarcinoma of the lung. Cancer
11. Grossi F, Loprevite M, Chiaramondia M, et al. Prognostic significance of K-ras, p53, bcl-2, PCNA, CD34 in radically resected non-small cell lung cancers. Eur J Cancer
12. Broermann P, Junker K, Brandt BH, et al. Trimodality treatment in Stage III nonsmall cell lung carcinoma: prognostic impact of K-ras mutations after neoadjuvant therapy. Cancer
13. Huncharek M, Muscat J, Geschwind JF. K-ras oncogene mutation as a prognostic marker in non-small cell lung cancer: a combined analysis of 881 cases. Carcinogenesis
14. Nelson HH, Christiani DC, Mark EJ, Wiencke JK, Wain JC, Kelsey KT. Implications and prognostic value of K-ras mutation for early-stage lung cancer in women. J Natl Cancer Inst
15. Mascaux C, Iannino N, Martin B, et al. The role of RAS oncogene in survival of patients with lung cancer: a systematic review of the literature with meta-analysis. Br J Cancer
16. Shepherd FA, Rodrigues Pereira J, et al. Erlotinib in previously treated non-small-cell lung cancer. N Engl J Med
17. Pao W, Miller VA. Epidermal growth factor receptor mutations, small-molecule kinase inhibitors, and non-small-cell lung cancer: current knowledge and future directions. J Clin Oncol
18. Pao W, Miller V, Zakowski M, et al. EGF receptor gene mutations are common in lung cancers from “never smokers” and are associated with sensitivity of tumors to gefitinib and erlotinib. Proc Natl Acad Sci USA
19. Lynch TJ, Bell DW, Sordella R, et al. Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-small-cell lung cancer to gefitinib. N Engl J Med
20. Paez JG, Janne PA, Lee JC, et al. EGFR mutations in lung cancer: correlation with clinical response to gefitinib therapy. Science
21. Eberhard DA, Johnson BE, Amler LC, et al. Mutations in the epidermal growth factor receptor and in KRAS are predictive and prognostic indicators in patients with non-small-cell lung cancer treated with chemotherapy alone and in combination with erlotinib. J Clin Oncol
22. Jackman DM, Yeap BY, Sequist LV, et al. Exon 19 deletion mutations of epidermal growth factor receptor are associated with prolonged survival in non-small cell lung cancer patients treated with gefitinib or erlotinib. Clin Cancer Res
23. Riely GJ, Pao W, Pham D, et al. Clinical course of patients with non-small cell lung cancer and epidermal growth factor receptor exon 19 and exon 21 mutations treated with gefitinib or erlotinib. Clin Cancer Res
24. Cappuzzo F, Ligorio C, Janne PA, et al. Prospective study of gefitinib in epidermal growth factor receptor fluorescence in situ hybridization-positive/phospho-Akt-positive or never smoker patients with advanced non-small-cell lung cancer: the ONCOBELL trial. J Clin Oncol
25. Takano T, Ohe Y, Sakamoto H, et al. Epidermal growth factor receptor gene mutations and increased copy numbers predict gefitinib sensitivity in patients with recurrent non-small-cell lung cancer. J Clin Oncol
26. Sequist LV, Joshi VA, Janne PA, et al. Response to treatment and survival of patients with non-small cell lung cancer undergoing somatic EGFR mutation testing. Oncologist
27. Paz-Ares L, Sanchez JM, Garcia-Velasco A, et al. A prospective phase II trial of erlotinib in advanced non-small cell lung cancer (NSCLC) patients (p) with mutations in the tyrosine kinase (TK) domain of the epirdermal growth facto receptor (EGFR). J Clin Oncol
28. Tsao MS, Sakurada A, Cutz JC, et al. Erlotinib in lung cancer—molecular and clinical predictors of outcome. N Engl J Med
29. Pao W, Ladanyi M, Miller VA. Erlotinib in lung cancer (Letter to the Editor). N Engl J Med
30. Herbst RS, Prager D, Hermann R, et al. TRIBUTE: a phase III trial of erlotinib hydrochloride (OSI-774) combined with carboplatin and paclitaxel chemotherapy in advanced non-small-cell lung cancer. J Clin Oncol
31. Kris MG, Natale RB, Herbst RS, et al. Efficacy of gefitinib, an inhibitor of the epidermal growth factor receptor tyrosine kinase, in symptomatic patients with non-small cell lung cancer: a randomized trial. JAMA
32. Johnson BE, Janne PA. Epidermal growth factor receptor mutations in patients with non-small cell lung cancer. Cancer Res
33. Pao W, Wang TY, Riely GJ, et al. KRAS mutations and primary resistance of lung adenocarcinomas to gefitinib or erlotinib. PLoS Med
34. Massarelli E, Varella-Garcia M, Tang X, et al. KRAS mutation is an important predictor of resistance to therapy with epidermal growth factor receptor tyrosine kinase inhibitors in non-small-cell lung cancer. Clin Cancer Res
35. Hirsch FR, Varella-Garcia M, Cappuzzo F, et al. Combination of EGFR gene copy number and protein expression predicts outcome for advanced non-small-cell lung cancer patients treated with gefitinib. Ann Oncol
36. Hirsch FR, Varella-Garcia M, Bunn PA Jr, et al. Epidermal growth factor receptor in non-small-cell lung carcinomas: correlation between gene copy number and protein expression and impact on prognosis. J Clin Oncol
37. Cappuzzo F, Hirsch FR, Rossi E, et al. Epidermal growth factor receptor gene and protein and gefitinib sensitivity in non-small-cell lung cancer. J Natl Cancer Inst
38. Pirker R, Filipits M, Dunant A, et al. IALT-Bio: a challenging research to improve adjuvant chemotherapy of completely resected NSCLC. J Thorac Oncol
39. Olaussen KA, Dunant A, Fouret P, et al. DNA repair by ERCC1 in non-small-cell lung cancer and cisplatin-based adjuvant chemotherapy. N Engl J Med
40. Zheng Z, Chen T, Li X, Haura E, Sharma A, Bepler G. DNA synthesis and repair genes RRM1 and ERCC1 in lung cancer. N Engl J Med
41. Bepler G, Sharma S, Cantor A, et al. RRM1 and PTEN as prognostic parameters for overall and disease-free survival in patients with non-small-cell lung cancer. J Clin Oncol
42. Davidson JD, Ma L, Flagella M, Geeganage S, Gelbert LM, Slapak CA. An increase in the expression of ribonucleotide reductase large subunit 1 is associated with gemcitabine resistance in non-small cell lung cancer cell lines. Cancer Res
43. Bepler G, Kusmartseva I, Sharma S, et al. RRM1 modulated in vitro and in vivo efficacy of gemcitabine and platinum in non-small-cell lung cancer. J Clin Oncol
44. Lord RV, Brabender J, Gandara D, et al. Low ERCC1 expression correlates with prolonged survival after cisplatin plus gemcitabine chemotherapy in non-small cell lung cancer. Clin Cancer Res
45. Ceppi P, Volante M, Novello S, et al. ERCC1 and RRM1 gene expressions but not EGFR are predictive of shorter survival in advanced non-small-cell lung cancer treated with cisplatin and gemcitabine. Ann Oncol
46. Cobo M, Isla D, Massuti B, et al. Customizing cisplatin based on quantitative excision repair cross-complementing 1 mRNA expression: a phase III trial in non-small-cell lung cancer. J Clin Oncol
47. Simon G, Sharma A, Li X, et al. Feasibility and efficacy of molecular analysis-directed individualized therapy in advanced non-small-cell lung cancer. J Clin Oncol
48. Filipits M, Pirker R, Dunant A, et al. Cell cycle regulators and outcome of adjuvant cisplatin-based chemotherapy in completely resected non-small-cell lung cancer: the International Adjuvant Lung Cancer Trial Biologic Program. J Clin Oncol
49. Esposito V, Baldi A, De Luca A, et al. Prognostic role of the cyclin-dependent kinase inhibitor p27 in non-small cell lung cancer. Cancer Res
50. Catzavelos C, Tsao MS, DeBoer G, Bhattacharya N, Shepherd FA, Slingerland JM. Reduced expression of the cell cycle inhibitor p27Kip1 in non-small cell lung carcinoma: a prognostic factor independent of Ras. Cancer Res
51. Hirabayashi H, Ohta M, Tanaka H, et al. Prognostic significance of p27KIP1 expression in resected non-small cell lung cancers: analysis in combination with expressions of p16INK4A, pRB, and p53. J Surg Oncol
2002;81:177–184, discussion 84.
52. Osoegawa A, Yoshino I, Tanaka S, et al. Regulation of p27 by S-phase kinase-associated protein 2 is associated with aggressiveness in non-small-cell lung cancer. J Clin Oncol
53. Seve P, Lai R, Ding K, et al. Class III beta-tubulin expression and benefit from adjuvant cisplatin/vinorelbine chemotherapy in operable non-small cell lung cancer: analysis of NCIC JBR. 10. Clin Cancer Res
54. Filipits M, Haddad V, Schmid K, et al. Multidrug resistance proteins do not predict benefit of adjuvant chemotherapy in patients with completely resected non-small cell lung cancer: International Adjuvant Lung Cancer Trial Biologic Program. Clin Cancer Res
55. Brambilla E, Soria J-C, Haddad V, et al. Prognostic and predictive value of apoptosis related factors Fas, FasL and survivin in non-small cell lung carcinoma patients enrolled in the IALT Trial. J Thorac Oncol
56. Potti A, Mukherjee S, Petersen R, et al. A genomic strategy to refine prognosis in early-stage non-small-cell lung cancer. N Engl J Med
57. Larsen JE, Pavey SJ, Passmore LH, Bowman RV, Hayward NK, Fong KM. Gene expression signature predicts recurrence in lung adenocarcinoma. Clin Cancer Res
58. Larsen JE, Pavey SJ, Passmore LH, et al. Expression profiling defines a recurrence signature in lung squamous cell carcinoma. Carcinogenesis
59. Beer DG, Kardia SL, Huang CC, et al. Gene-expression profiles predict survival of patients with lung adenocarcinoma. Nat Med
60. Wigle DA, Jurisica I, Radulovich N, et al. Molecular profiling of non-small cell lung cancer and correlation with disease-free survival. Cancer Res
61. Guo L, Ma Y, Ward R, Castranova V, Shi X, Qian Y. Constructing molecular classifiers for the accurate prognosis of lung adenocarcinoma. Clin Cancer Res
62. Raponi M, Zhang Y, Yu J, et al. Gene expression signatures for predicting prognosis of squamous cell and adenocarcinomas of the lung. Cancer Res
63. Chen HY, Yu SL, Chen CH, et al. A five-gene signature and clinical outcome in non-small-cell lung cancer. N Engl J Med
64. Lu Y, Lemon W, Liu PY, et al. A gene expression signature predicts survival of patients with stage I non-small cell lung cancer. PLoS Med
65. Lau SK, Boutros PC, Chang QZ, et al. A six-gene prognostic classifier of early-stage non-small cell lung carcinoma. In Proceedings of the 97th Annual Meeting of the American Association of Cancer Research, Washington D.C. 2006(abstract 3379).
66. Dupuy A, Simon RM. Critical review of published microarray studies for cancer outcome and guidelines on statistical analysis and reporting. J Natl Cancer Inst
67. Massague J. Sorting out breast-cancer gene signatures. N Engl J Med
68. Febbo PG, Kantoff PW. Noise and bias in microarray analysis of tumor specimens. J Clin Oncol
69. Lin DW, Coleman IM, Hawley S, et al. Influence of surgical manipulation on prostate gene expression: implications for molecular correlates of treatment effects and disease prognosis. J Clin Oncol
70. Bild AH, Yao G, Chang JT, et al. Oncogenic pathway signatures in human cancers as a guide to targeted therapies. Nature
71. Pan Q, Pao W, Ladanyi M. Rapid polymerase chain reaction-based detection of epidermal growth factor receptor gene mutations in lung adenocarcinomas. J Mol Diagn
72. Tang K, Fu DJ, Julien D, Braun A, Cantor CR, Koster H. Chip-based genotyping by mass spectrometry. Proc Natl Acad Sci USA
73. Miller VA, Krug LM, Ng KK, et al. Phase II trial of docetaxel and vinorelbine in patients with advanced non-small-cell lung cancer. J Clin Oncol
74. Page RD, Smith FP, Geils GF, Beall CL, Fridman M, Allen BJ. Dose-dense vinorelbine and docetaxel with Filgrastim support in patients with advanced non-small cell lung carcinoma. Cancer
75. Lu Y, Bellgrau D, Dwyer-Nield LD, et al. Mutation-selective tumor remission with Ras-targeted, whole yeast-based immunotherapy. Cancer Res
76. Niedernhofer LJ, Bhagwat N, Wood RD. ERCC1 and Non-Small Cell Lung Cancer (Letter to the Editor). N Engl J Med
77. Pao W, Ladanyi M. Epidermal growth factor receptor mutation testing in lung cancer: searching for the ideal method. Clin Cancer Res
Copyright © 2008 by the European Lung Cancer Conference and the International Association for the Study of Lung Cancer.