Histologic Subtype in NSCLC: Does It Matter?Published on Physicians Practice (http://www.physicianspractice.com)

Histologic Subtype in NSCLC: Does It Matter?November 25, 2009By Giovanni Selvaggi, MD [1] and Giorgio V. Scagliotti, MD, PhD [2]

Since the publication of a meta-analysis in 1995 that demonstrated a modest survival benefitcompared to best supportive care, platinum-based chemotherapy became the cornerstone oftherapy in the first-line setting in advanced-stage non–small-cell lung cancer (NSCLC) for patientswith good performance status.[1] A recent meta-analysis of 16 randomized trials including 2,714patients demonstrated an advantage of chemotherapy over best supportive care with an absoluteimprovement in survival of 9% at 12 months.[2]

Since the publication of a meta-analysis in 1995 that demonstrated a modest survival benefitcompared to best supportive care, platinum-based chemotherapy became the cornerstone oftherapy in the first-line setting in advanced-stage non–small-cell lung cancer (NSCLC) for patientswith good performance status.[1] A recent meta-analysis of 16 randomized trials including 2,714patients demonstrated an advantage of chemotherapy over best supportive care with an absoluteimprovement in survival of 9% at 12 months.[2]

Therapeutic guidelines indicate doublets of cisplatin or carboplatin with either gemcitabine(Gemzar), taxanes (paclitaxel or docetaxel [Taxotere]), or vinorelbine as reference regimens.[3,4]These guidelines are based on randomized phase III studies that compared these chemotherapycombinations “head to head” without detecting relevant differences in efficacy.[5-7] However,evidence from several randomized clinical trials and systematic reviews suggest that cytotoxicchemotherapy has reached an efficacy plateau.[8,9]

Currently the optimal therapeutic strategy for the individual patient diagnosed with stage IIIB or IVNSCLC is quite difficult to define since the decision-making process is shifting very quickly toward anindividualized approach. Emerging evidence suggests that aside from the effects of well-knownclinical parameters such as stage of disease and patient performance status, lung cancer behavesdifferently in different patients as the result of a variety of molecular profiles. The impact of histologyin this context has been overlooked in the past.The role of histology in addition to tumor-node-metastasis (TNM) stage was assessed in surgicallymanaged stage I–IIIA NSCLC cases from the database of the International Association for the Study ofLung Cancer (IASLC), in the process of defining the new TNM staging system for lung cancer in therecently released 7th edition of the International Union Against Cancer and American JointCommittee on Cancer’s TNM Classification of Malignant Tumors. In 12,428 cases of NSCLC, age,gender, and performance status—but not histology—were found to be prognostic for survival afteradjustment for stage of disease. Adenocarcinoma was more frequent than squamous cell carcinoma(SCC) in women (55% vs 25%), while the opposite was true in men (30% vs 57%). In a modelincluding histology, pathologic stage, gender, and age, SCC patients had a significant survivaladvantage over those with adenocarcinoma or large-cell carcinoma (LCC), with no significant

Page 1 of 10

Histologic Subtype in NSCLC: Does It Matter?Published on Physicians Practice (http://www.physicianspractice.com)

difference between LCC and adenocarcinoma. Cases designated as bronchioloalveolar carcinoma(BAC) had a superior prognosis compared with other histotypes. The better prognosis for SCCcompared with other histotypes could not be demonstrated in women.[10]In the same way, the Japanese Joint Committee of Lung Cancer Registry reported over 13,000 lungcancer cases that underwent surgery in 2002. They found a better prognosis for women andadenocarcinoma, with a 67% 5-year survival for adenocarcinoma vs 53% for SCC. In this study, theanalysis of prognostic factors (including histology) was not adjusted for stage or gender; suchadjustments could have led to different results.[11] In fact, by adjusting for stage and gender, anyeffect of imbalance that would favor adenocarcinoma is diminished, thus revealing a possiblesurvival advantage for SCC.Histology of Lung Cancer

Currently, the World Health Organization (WHO) classification of lung tumors should remain thecommon ground by which to compare results of studies potentially aimed at identifying a differentialactivity of any cytotoxic regimen according to histology.In the past, oncologists dealing with lung cancer routinely distinguished only between small-cell lungcancer (SCLC) and NSCLC to select for the appropriate therapeutic management. NSCLCencompasses several subtypes with different morphologic features that are often treated accordingto similar strategies, so in everyday practice a clear-cut distinction among such histotypes was notconsidered mandatory.[12] Conversely, the WHO classifications of lung cancer have always kept thedifferent histologic types separated, with individual categories for SCC, adenocarcinoma, andLCC.[13] Over the past few decades, the changing habits of cigarette smoking and enviromentalpollution have contributed to changes in the relative frequency of these histologic subtypes.Adenocarcinoma shows a growing incidence, particularly in younger nonsmoking females, whereasSCC and SCLC are on the decline.[14]The histologic recognition of classical SCC is generally straightforward, with distinctive areas ofkeratinization and an associated inflammatory component, especially in cases undergoing cavitation.Less differentiated forms of SCC, without keratinization and with smaller cells, may resemble basalcell layers of the squamous epithelium. In these cases, immunohistochemistry (IHC) may be requiredto detect specific types of cytokeratins, and neuroendocrine markers may also be helpful.Adenocarcinoma includes a morphologically heterogeneous group of tumors that are less frequentlyassociated with tobacco smoking than SCC. Adenocarcinomas are made of glands with papillarystructures, or solid growth with mucin production.BAC is still classified as an adenocarcinoma, although its pathologic definition has undergone majorchanges in recent years. Indeed, newly proposed strict criteria for the diagnosis of BAC are notcorrectly applied across all institutions, and when these criteria are followed, pure BAC casesbecome quite rare. The diagnosis of BAC is now restricted to tumors that have a lepidic growth alongpulmonary septa in the absence of parenchymal invasion,[15] with further division into mucinousand nonmucinous forms. In clinical practice, however, the most common finding is adenocarcinomashowing a combined pattern of growth, with conventional acinar or papillary adenocarcinomaassociated with mixed areas of BAC in the same tumor. Such tumors are classified as a mixedsubtype of adenocarcinoma.

LCC is the least frequent variant of NSCLC, and it suffers from the lack of standardized morphologicdiagnostic criteria. LCC underwent major changes in the last classification scheme, withundifferentiated pleomorphic and/or sarcomatoid variants moved to a new group designated as“sarcomatoid carcinoma.” The heading of LCC includes rarer lung cancer types, such as basaloid,lymphoepithelial, clear-cell, rhabdoid, and the large-cell neuroendocrine carcinoma (LCNEC). Mostcases without clear signs of glandular or squamous differentiation are reported as large-cell(anaplastic) carcinoma . Most undifferentiated cases of LCC should probably be reclassified asadenocarcinoma or SCC on the basis of their molecular features. LCNEC should be identified usingneuroendocrine markers and should be excluded from studies in NSCLC because of its differentbiologic and clinical properties.Technical Problems Dealing With Limited Pathologic Specimens

A critical step is reached when considering the amount and the quality of the specimen that isreceived by the pathologist. In fact, diagnostic procedures routinely performed in thoracic oncologyto obtain pathologic tissue tend to become increasingly less invasive. Nonetheless, cytology or smallbiopsies often yield a limited amount of viable cells or tissue, therefore allowing even an experienced

Page 2 of 10

Histologic Subtype in NSCLC: Does It Matter?Published on Physicians Practice (http://www.physicianspractice.com)

pathologist to formulate a generic diagnosis of NSCLC–not otherwise specified (NOS).[16] In all theseundifferentiated cases, IHC can assist pathologists in fine-tuning the diagnosis. Currently, markerssuch as thyroid transcription factor-1 (TTF-1), napsin and cytokeratin 7 (mainly for adenocarcinoma),and p63 and cytokeratin 5 (mainly for SCC) are more frequently used in the pathologic diagnosticprocess to trace signs of lineage differentiation.Major advances in the phenotypic profiling of lung cancer were obtained by molecular analysis of alarge number of genes, which provide different biologic signatures for each histologic subtype. As aresult, a recent study described four genes that were upregulated in SCC by at least 20-fold incomparison to adenocarcinoma and/or normal lung parenchyma (PKP1, desmocollin-3 [DSC3], p63,and CK17).[17] The greatest difference between SCC and adenocarcinoma (and also between tumorand normal tissues) was observed for DSC3, a protein localized in desmosomal junctions of stratifiedepithelia. Gene-expression profiling data confirmed a differential expression of DSC3 in NSCLC.[18]DSC3 was expressed in almost half of a sample of undifferentiated LCC and was mutually exclusivewith TTF-1, definining residual squamous differentiation of cells. Coexpression of the two markers isrestricted to adenosquamous carcinomas. Thus, a combined morphologic-phenotypic approach mayrepresent a valid diagnostic tool in lung cancer, as we move toward tailoring chemotherapyaccording to histology. Similarly, high-throughput microarrays have been used to measure microRNAexpression levels in 122 adenocarcinoma and SCC samples and have identified hsa-miR-205 as ahighly specific marker for SCC. Upon standardization, this diagnostic assay could provide highlyaccurate subclassification of NSCLC patients.[19]The reproducibility of squamous vs nonsquamous classification could still largely be ameliorated. Astudy of 96 primary lung tumors showed that reproducible diagnosis of SCC based on morphologyalone is inadequate. As expected in the era of histology-guided therapy, this highlights the need forstricter diagnostic criteria and confirmatory IHC stains in the diagnosis of SCC.[20]Prognostic Impact of Histology

In NSCLC, tumor histology has not been consistently identified as a prognostic factor independent oftreatment choice. Clinical studies evaluating the efficacy of chemotherapy in advanced NSCLC over a25-year period have been reviewed, and a weak association between histology and therapeuticoutcome has emerged, with a better median survival in patients with a nonsquamous histology.Indeed, none of the studies was planned to include a test of treatment-by-histology interaction. In 11studies, an association between histology and prognosis was reported, while in 7 studies histologicsubtype was even predictive of outcome after a specific chemotherapy regimen.

When evaluating trials of targeted therapies, 14 publications reported that histology was prognosticand/or predictive in patients treated with epidermal growth factor receptor (EGFR) tyrosine-kinaseinhibitors (TKIs). A small study showed a nonsignificant trend to higher efficacy of cisplatin andetoposide vs a four-drug regimen in patients with SCC (response rate = 44% vs 21%, respectively)and not in patients with adenocarcinoma. Similarly, in a phase III study, nonsquamous histology wasprognostic in patients treated with chemotherapy compared with best supportive care, while in

Page 3 of 10

Histologic Subtype in NSCLC: Does It Matter?Published on Physicians Practice (http://www.physicianspractice.com)

patients with squamous histology no difference in outcome was seen. In a Greek study, patients withadenocarcinoma had a significantly greater response to gemcitabine plus docetaxel, whereaspatients with SCC had a higher response rate to cisplatin plus docetaxel, but no differences werefound in progression-free survival (PFS) and overall survival (OS).[21]In an Italian phase III trial in advanced NSCLC testing three platinum-based doublets, no significanttreatment-by-histology interaction was seen for survival. However, when pairwise comparisons ofhistology groups were analyzed, a statistically significant survival advantage in patients with SCCover adenocarcinoma became evident.Recently, an individual patient data meta-analysis considered nine randomized clinical trialsincluding almost 3,000 patients who were treated with doublets of cisplatin or carboplatin and athird-generation agent. Statistically significant tests of treatment-by-histology interaction confirmedthe predictive effect of histology on survival. In nonsquamous cases, carboplatin-containingregimens were associated with lower odds of response and lower survival rates.[22]These studies show how histology never really emerged as a strong prognostic factor that couldpredict response and/or survival. Hints of small differences suggesting a differential behavior of thetumor in relation to chemotherapy according to histology were there to be interpreted in laterstudies.Differential Efficacy of Chemotherapy According to Histology

Until recently, only the coarse distinction between NSCLC and SCLC had relevant therapeuticimplications. The doublet combination of platinum and etoposide is still unbeaten as the first-lineapproach in SCLC but is not used anymore in NSCLC. Historically, within the NSCLC subtypes therewas no evidence of superior efficacy for a specific cytotoxic drug administered as a single agent oras part of a combination. Based on the results of randomized clinical trials, each institution ormedical oncologist has been choosing a doublet, mainly based on local policies, costs, orconvenience of the therapeutic schedule.

Recently, however, new findings prompted a change in clinical practice. The therapeutic advantageof pemetrexed (Alimta) and cisplatin compared to gemcitabine and cisplatin in nonsquamousNSCLC[23] and the safety issues regarding the use of bevacizumab (Avastin) in squamousNSCLC[24] brought up histology as a new evidence-based selection criterion for choosing the mostappropriate chemotherapy regimen. The first study was a phase III multicenter trial including 1,725chemotherapy-naive patients with advanced NSCLC who were randomized to receive cisplatin andpemetrexed or cisplatin and gemcitabine (Table 1). The trial was planned according to anoninferiority study design with OS as a primary endpoint. Preplanned subset analyses were carriedout to evaluate differential efficacy according to histology. Patients with adenocarcinoma (847patients) and with LCC (153 patients) had a longer OS (12.6 vs 10.9 months and 10.4 vs 6.7 months,respectively) when they were treated with cisplatin and pemetrexed. Conversely, the 473 patientswith SCC had a survival advantage when they received cisplatin and gemcitabine (10.8 vs 9.4months). A significant treatment-by-histology interaction was observed for both PFS and OS. As aresult, pemetrexed received worldwide approval for use in combination with cisplatin for first-linetreatment in patients with locally advanced or metastatic nonsquamous NSCLC.The same trend was seen in a retrospective analysis from a second-line study of pemetrexed vsdocetaxel.[25] Patients with nonsquamous histology treated with pemetrexed had a statisticallysignificant superior survival compared to those treated with docetaxel (hazard ratio [HR] = 0.778,95% confidence interval [CI] = 0.607–0.997), whereas in patients with SCC, docetaxel was superiorto pemetrexed (HR = 1.563, 95% CI = 1.079–2.264). The investigators found a significanttreatment-by-histology interaction. In addition, a phase II randomized study conducted in Japancompared two doses (500 vs 900 mg) of single-agent pemetrexed as second- or third-line therapy inadvanced NSCLC and also showed that efficacy varied according to histology.[26]Pemetrexed was also tested in a phase III study in comparison to placebo as maintenance therapy inadvanced NSCLC that had not progressed after four cycles of front-line platinum-based standardchemotherapy. PFS was the primary endpoint of the study and was significantly longer withpemetrexed (4 vs 2 months).Overall survival was also significantly better in the pemetrexed arm (13.4 vs 10.6 months). Theimprovements in PFS and OS were more striking in adenocarcinoma (PFS = 4.4 vs 1.8 months, HR =0.47; OS = 15.5 vs 10.3 months; HR = 0.70), while pemetrexed did not improve either outcomeparameter in SCC.[27]The bulk of data coming from the three above mentioned studies supports a higher efficacy of

Page 4 of 10

Histologic Subtype in NSCLC: Does It Matter?Published on Physicians Practice (http://www.physicianspractice.com)

pemetrexed in adenocarcinoma, either as first-line or second-line therapy or as a maintenancestrategy. A potential explanation for this biologic phenomenon relies in the mechanism of action ofpemetrexed. Thymidylate synthase (TS) is the main cellular target of pemetrexed, and higher levelsof TS promote resistance to pemetrexed action, as seen in a variety of solid tumors.[28] In lungcancer, a differential expression of TS among histotypes has been extensively described. Baseline TSgene and protein expressions are significantly higher in SCC compared to adenocarcinoma.[29] Inaddition, TS and S phase kinase–associated protein (Skp2) are transcriptionally regulated in the Sphase of the cell cycle by the transcription factor E2F1, and elevated expression of Skp2 has beenfound more commonly in patients with SCC than in those with adenocarcinoma.[30]In LCC, significantly higher median TS levels were found compared to adenocarcinoma, but nocorrelation was found between TS and either Ki-67 or E2F1. Also in LCC, significantly higher TS levelswere observed in DSC3-positive tumors (resembling SCC differentiation) compared to DSC3-negativetumors, and lower TS levels were seen in TTF-1 positive LCC (resembling adenocarcinomadifferentiation).[31] Finally, the highest levels of TS gene and protein expression were registered inSCLC.[32] A recent phase III study in extensive-disease SCLC, which investigated the combination ofpemetrexed and carboplatin vs etoposide and carboplatin, was stopped early due to the lack ofefficacy seen in the pemetrexed arm.[33]Histology and Targeted Therapies

Advances in the understanding of the molecular biology of cancer have made possible the discoveryof several molecular targets and the consequent development of novel therapies with the primarygoal of switching off signaling pathways involved in the uncontrolled growth of cancer cells.Inhibition of EGFR signaling and blockade of angiogenesis have been identified as key therapeutictargets in NSCLC. Strategies to block these pathways mainly include TKIs and monoclonal antibodies.Erlotinib (Tarceva) and gefitinib (Iressa) are the two EGFR-TKIs with proven efficacy against NSCLC.Recent phase II studies have shown that both agents could be more active than chemotherapy asfirst-line regimens in advanced NSCLC when the population is selected for the presence of EGFRmutations or EGFR gene amplification,[34] as detected by fluorescence in situ hybridization (FISH). In2009, two papers reported about the use of TKIs as first-line therapy in selected patients withadvanced NSCLC. Both studies showed significant improvements in response rate and PFS inpatients whose tumors had EGFR mutations. A Spanish group screened over 2,000 patients for thepresence of EGFR mutations, which were found in 16% of cases and, as expected, were morefrequent in women, adenocarcinomas, and nonsmoking patients.[35] Patients who presented with anEGFR mutation were treated with upfront erlotinib, and had a median survival of 27 months.In the Iressa Pan-Asia Study (IPASS), gefitinib was compared to carboplatin/paclitaxel as first-linetreatment in 1,217 never (or light) smokers with Asian ethnicity and a histologic diagnosis ofadenocarcinoma.[36] This study demonstrated the superiority of gefitinib relative to thecarboplatin/paclitaxel doublet in terms of PFS (12-month PFS = 25% vs 7%). EGFR gene mutationswere assessed in 36% of all randomized patients. The benefit was largely confined to patientsharboring that mutation, and it may not depend strictly on histology. Overall survival was notprolonged by gefitinib, and this remains a key issue to be addressed by future trials.New results from the phase III Sequential Tarceva in Unresectable NSCLC (SATURN) trial presentedat the 13th World Conference on Lung Cancer, organized by the IASLC, show that erlotinibmaintenance therapy after first-line platinum-based doublets in almost 889 patients with advancednonprogressing NSCLC significantly improves OS (12 vs 11 months).[37] Survival was a secondaryendpoint of this trial. PFS, the primary endpoint, was increased 41% to 45% by erlotinib. Subgroupanalyses for clinical predictive factors did not show any difference in outcome. Unexpectedly, allpatients—regardless of sex, histology, and nonsmoking status—derived some benefit from the drug.The investigators found a 10% absolute difference in survival between the treatment and placebogroups at 3 years. However, the most consistent improvement of outcome parameters was recordedin the small subset of patients with EGFR mutation.Clinical parameters may help the clinician to foresee sensitivity to EGFR TKIs. Female sex, Asianethnicity, nonsmoker status, and adenocarcinomatous histology predicted for the presence of EGFRmutations and, therefore, can be considered clinical predictors of response.[38] In the end, EGFRmutations are mostly present in adenocarcinoma rather than SCC, but the future wider availability ofkits to test for EGFR mutations will overcome the need to choose EGFR TKIs vs cytotoxicchemotherapy based solely on histotype. Conversely, such clinical parameters did not apply to themonoclonal antibody cetuximab (Erbitux) in the First-Line Erbitux in Lung Cancer (FLEX) trial, wherecetuximab was added to the combination of cisplatin and vinorelbine and prolonged survival

Page 5 of 10

Histologic Subtype in NSCLC: Does It Matter?Published on Physicians Practice (http://www.physicianspractice.com)

compared to chemotherapy alone.[39] Data do not support the hypothesis that KRAS mutationstatus is predictive for cetuximab efficacy when combined with chemotherapy, as seen previously incolon carcinoma.[40] Only the emergence of early cutaneous rash was predictive for outcome.[41]Patients treated with cetuximab who developed an early acne-like rash of any grade had a longer

median OS than those without the rash.A new molecular subset of NSCLC is defined by the expression of EML4-ALK. Patients most likely toharbor the EML4-ALK fusion oncogene are young, mainly male, never/light smokers withadenocarcinoma. Because some of these features are also associated with EGFR mutation, it isessential to screen such patients by mutation testing and not to rely solely on the presence ofclinical predictors. EML4-ALK expression is mutually exclusive to EGFR and KRAS mutations.[42]EML4-ALK allows a step further in defining newer subsets of adenocarcinoma based on genomicprofiling. They represent probably around 5% of NSCLC patients. Those who harbor this mutation donot benefit from EGFR TKIs and should be directed to trials with ALK-targeted inhibitors.Coming back to the other disrupted pathway in lung cancer, it is well recognized thatneoangiogenesis, the formation of new blood vessels, is a fundamental process for the growth ofsolid tumors and their metastatic spread. Bevacizumab is a recombinant, humanized, monoclonalantibody against vascular endothelial growth factor (VEGF), and is currently the most extensivelyinvestigated antiangiogenic agent.

Page 6 of 10

Histologic Subtype in NSCLC: Does It Matter?Published on Physicians Practice (http://www.physicianspractice.com)

In a phase II study of bevacizumab and chemotherapy, fatal hemorrhagic episodes turned out to bemore frequent when central cavitated lesions were present, as is more frequently observed inSCC.[24] Therefore, squamous histology, metastases to the central nervous system, history ofhemoptysis, and history of documented hemorrhagic diathesis became exclusion criteria for phase IIIstudies. Once again, histotype is a potential limitation in the use of a new drug, with bevacizumabuse favored for adenocarcinoma.In a phase III trial (Eastern Cooperative Oncology Group [ECOG] 4599), the combination ofbevacizumab and paclitaxel/carboplatin followed by bevacizumab until disease progression showedan improvement in overall response rate (35% vs 15%), PFS (6.2 vs 4.5 months), and OS (12.3 vs10.3 months) over paclitaxel/carboplatin alone (Table 2).[43] A pan-European study following asimilar design but with cisplatin/gemcitabine as the chemotherapy backbone met the primaryendpoint of improving PFS but did not confirm any survival gain,[44] probably due to poststudytherapies administered when disease progressed. In both studies, greater toxicity was seen forpatients receiving bevacizumab, especially in terms of hypertension, proteinuria, bleeding,neutropenia, and thrombocytopenia. A recent multicenter phase II study of pemetrexed andcarboplatin plus bevacizumab followed by maintenance pemetrexed and bevacizumab demonstratedencouraging activity in patients with chemotherapy-naive advanced nonsquamous NSCLC (responserate of 55% and OS of 14 months).[45]A new class of targeted agents is represented by inhibitors of insulin-like growth factor type 1receptor (IGF-IR). IGF-IR is often overexpressed in lung tumors and mediates the proliferation of lungcancer cells and their resistance to therapy. It has been reported to be expressed at higher levels inSCC than in adenocarcinoma. The IGF-IIR gene is mutated in approximately 60% of SCC cases and27% of adenocarcinomas.[46] Therefore, a promising strategy for inhibiting the function of IGF-IR isthe use of antibodies that bind to the extracellular domain of this receptor.CP-751,871 (figitumumab) is a fully human monoclonal antibody that proved active in NSCLC whenassociated with carboplatin/paclitaxel treatment in a phase II study conducted inchemotherapy-naive patients.[47] A tendency toward higher response rates (70%) was observed inSCC compared to adenocarcinoma with the addition of CP-751,871 to carboplatin and paclitaxel,whereas NSCLC-NOS patients (almost one-third of those studied) did not receive any additionalbenefit. Such data reflect a differential expression of IGF-IR across NSCLC subtypes and, once more,highlight the need for a more precise histologic definition.Conclusions

NSCLC results from a multistep process linked to several intracellular pathways and as a sum ofmultiple genetic alterations. Data from available randomized studies have yielded sufficient evidencethat chemosensitivity varies consistently according to histotype. Even within a morphologicallydefined histologic subtype such as adenocarcinoma, multiple subtypes can exist, each associatedwith a different prognosis and/or responsiveness to a specific drug. In a subset of adenocarcinoma ofthe lung, positivity for alpha and beta estrogen receptor expression was associated with distinctiveclinicopathologic and genetic features. Alpha estrogen receptor expression correlated with EGFRmutations and represented a worse prognostic factor.[48]Future research should be aimed at exploiting combinations of drugs to target different pathways atthe same time, while implementing a central pathology review (or other related methods) to verifyhistologic diagnosis and improve trial designs based on histology. References: 1. Non-small Cell Lung Cancer Collaborative Group: Chemotherapy in non-small celllung cancer: A meta-analysis using updated data on individual patients from 52 randomised clinicaltrials. BMJ 311:899-909, 1995.2. NSCLC Meta-Analyses Collaborative Group: Chemotherapy in addition to supportive care improvessurvival in advanced non-small-cell lung cancer: A systematic review and meta-analysis of individualpatient data from 16 randomized controlled trials. J Clin Oncol 26:4617-4625, 2008.3. Clinical practice guidelines for the treatment of unresectable non-small-cell lung cancer. Adoptedon May 16, 1997 by the American Society of Clinical Oncology. J Clin Oncol 15:2996-3018, 1997.4. D’Addario G, Felip E, for the ESMO Guidelines Working Group: Non-small-cell lung cancer: ESMOclinical recommendations for diagnosis, treatment and follow-up. Ann Oncol 20(suppl 4):68-70, 2009.5. Scagliotti GV, De Marinis F, Rinaldi M, et al: Phase III randomized trial comparingthree-platinum-based doublets in advanced non-small-cell lung cancer. J Clin Oncol 20:4285-4291,2002.

Page 7 of 10

Histologic Subtype in NSCLC: Does It Matter?Published on Physicians Practice (http://www.physicianspractice.com)

6. Kelly K, Crowley J, Bunn PA Jr, et al: Randomized phase III trial of paclitaxel plus carboplatin versusvinorelbine plus cisplatin in the treatment of patients with advanced non–small-cell lung cancer: ASouthwest Oncology Group trial. J Clin Oncol 19:3210-3218, 2001.7. Fossella F, Pereira JR, von Pawel J, et al: Randomized, multinational, phase III study of docetaxelplus platinum combinations versus vinorelbine plus cisplatin for advanced non-small-cell lung cancer:The TAX 326 study group. J Clin Oncol 21:3016-3024, 2003.8. Delbaldo C, Michiels S, Rolland E, et al: Second or third additional chemotherapy drug fornon-small cell lung cancer in patients with advanced disease. Cochrane Database Syst Rev(4):CD004569, 2007.9. Breathnach OS, Freidlin B, Conley B, et al: Twenty-two years of phase III trials for patients withadvanced non-small-cell lung cancer: Sobering results. J Clin Oncol 19:1734-1742, 2001.10. Sculier J-P, Chansky K, Crowley J, et al: The impact of additional prognostic factors on survivaland their relationship with the anatomical extent of disease expressed by the 6th Edition of the TNMClassification of Malignant Tumors and the proposals for the 7th Edition. J Thorac Oncol 3:457-466,2008.11. Asamura H, Goya T, Koshiishi Y, et al: Japanese Joint Committee of Lung Cancer Registry. AJapanese lung cancer registry study. Prognosis of 13,010 resected lung cancers. J Thorac Oncol3:46-52, 2008.12. Hirsch FR, Speafico A, Novello S, et al: The prognostic and predictive role of histology inadvanced non-small cell lung cancer: A literature review. J Thorac Oncol 3:1468-1481, 2008.13. Travis WD, Brambilla E, Muller-Hermelink HK, et al: Pathology and genetics of tumors of the lung,pleura, thymus and heart, in WHO Classification of Tumours, pp 9-124. IARC Press, Lyon, 2004.14. Alberg AJ, Ford JG, Samet JM: Epidemiology of lung cancer: ACCP evidence-based clinical practiceguidelines, 2nd ed. Chest 132:29S-55S, 2007.15. Travis WD, Garg K, Franklin WA, et al: Evolving concepts in the pathology and computedtomography imaging of lung adenocarcinoma and bronchioloalveolar carcinoma. J Clin Oncol23:3279-3287, 2005.16. Cataluna JJ, Perpina M, Greses JV, et al: Cell type accuracy of bronchialbiopsy specimens inprimary lung cancer. Chest 109:1199-1203, 1996.17. Angulo B, Suarez-Gauthier A, Lopez-Rios F, et al: Expression signatures in lung cancer reveal aprofile for EGFR-mutant tumours and identify selective PIK3CA overexpression by gene amplification.J Pathol 214:347-356, 2008.18. Monica V, Ceppi P, Righi L, et al: Further subtyping of undifferentiated large cell carcinoma of thelung at the advent of histotype-specific chemotherapy. Modern Pathol 22:709-717, 2009.19. Lebanony D, Benjamin H, Gilad S, et al: Diagnostic assay based on hsa-miR-205 expressiondistinguishes squamous from nonsquamous non-small-cell lung carcinoma. J Clin Oncol27:2030-2037, 2009.20. Grilley-Olson JE, Hayes DN, Qaqish BF, et al: Diagnostic reproducibility of squamous cellcarcinoma (SC) in the era of histology-directed non-small cell lung cancer (NSCLC) chemotherapy: Alarge prospective study (abstract 8008). J Clin Oncol 27(15S):409s, 2009.21. Georgoulias V, Papadakis E, Alexopoulos A, et al: Platinum-based and non-platinum-basedchemotherapy in advanced non-small-cell lung cancer: A randomised multicentre trial. Lancet357:1478-1484, 2001.22. Ardizzoni A, Boni L, Tiseo M, et al, for the CISCA (CISplatin versus CArboplatin) Meta-analysisGroup: Cisplatin- versus carboplatin-based chemotherapy in first-line treatment of advancednon-small-cell lung cancer: An individual patient data meta-analysis. J Natl Cancer Inst 99:847-857,2007.23. Scagliotti GV, Parikh P, von Pawel J, et al: Phase III study comparing cisplatin plus gemcitabinewith cisplatin plus pemetrexed in chemotherapy-naive patients with advanced-stage non-small-celllung cancer. J Clin Oncol 26:3543-3551, 2008.24. Johnson DH, Fehrenbacher L, Novotny WF, et al: Randomized phase II trial comparingbevacizumab plus carboplatin and paclitaxel with carboplatin and paclitaxel alone in previouslyuntreated locally advanced or metastatic non-small-cell lung cancer. J Clin Oncol 22:2184-2191,2004.25. Peterson P, Park K, Fossella F, et al: Is pemetrexed more effective in adenocarcinoma and largecell lung cancer than in squamous cell carcinoma? A retrospective analysis of a phase III trial ofpemetrexed vs docetaxel in previously treated patients with advanced non-small cell lung cancer(NSCLC). J Thorac Oncol 2:S851, 2007.26. Okabe T, Kubota K, Tamura T, et al: Prognostic factors affecting survival on pretreated patients

Page 8 of 10

Histologic Subtype in NSCLC: Does It Matter?Published on Physicians Practice (http://www.physicianspractice.com)

with locally advanced or metastatic non-small cell lung cancer (NSCLC)—subgroup analysis in arandomized Ph II study of pemetrexed 500 mg/m2 and 1000 mg/m2. Eur J Cancer Suppl 5(4):376,2007.27. Belani CP, Brodowicz Y, Ciuleanu T , et al: Maintenance pemetrexed (Pem) plus best supportivecare (BSC) versus placebo (Plac) plus BSC: A randomized phase III study in advanced non-small celllung cancer (NSCLC) (abstract CRA8000). J Clin Oncol 27(18S), 2009.28. Shih C, Chen VJ, Gossett LS, et al: LY231514, a pyrrolo[2,3-d]pyrimidine-based antifolate thatinhibits multiple folate-requiring enzymes. Cancer Res 57:1116-1123, 1997.29. Ceppi P, Volante M, Saviozzi S, et al: Squamous cell carcinoma of the lung compared with otherhistotypes shows higher messenger RNA and protein levels for thymidylate synthase. Cancer107:1589-1596, 2006.30. Salon C, Merdzhanova G, Brambilla C, et al: E2F-1, Skp2 and cyclin E oncoproteins areupregulated and directly correlated in high-grade neuroendocrine lung tumors. Oncogene26:6927-6936, 2007.31. Scagliotti GV, Monica V, Ceppi P, et al: Baseline thymidylate synthase expression according tohistological subtypes of non-small cell lung cancer (abstract 7521). J Clin Oncol 27(15S):387s, 2009.32. Ceppi P, Volante M, Ferrero A, et al: Thymidylate synthase expression in gastroenteropancreaticand pulmonary neuroendocrine tumors. Clin Cancer Res 14:1059-1064, 2008.33. Smit F, Socinski MA, Mullaney BP, et al: Pharmacogenomic analysis from a phase III study ofpemetrexed plus carboplatin (PC) versus etoposide plus carboplatin (EC) in chemonaive patients(pts) with extensive-stage disease small cell lung cancer (ED-SCLC) (abstract 8030). J Clin Oncol27(15S):414s, 2009.34. Hirsch FR, Varella-Garcia M, Bunn PA Jr, et al: Molecular predictors of outcome with gefitinib in aphase III placebo-controlled study in advanced non-small-cell lung cancer. J Clin Oncol 24:5034-5042,2006.35. Rosell R, Moran T, Queralt C, et al: Screening for epidermal growth factor receptor mutations inlung cancer. N Engl J Med 361:958-967, 2009.36. Mok TS, Wu Y-L, Thongprasert S, et al: Gefitinib or carboplatin–paclitaxel in pulmonaryadenocarcinoma. N Engl J Med 361:947-957, 2009.37. Cappuzzo F, Coudert B, Wierzbicki R, et al: Efficacy and safety of erlotinib as first-linemaintenance in NSCLC following non-progression with chemotherapy: Results from the phase IIISATURN study (abstract A2.1). J Thorac Oncol 4(S1), 2009.38. Sequist LV, Bell DW, Lynch TJ, et al: Molecular predictors of response to epidermal growth factorreceptor antagonists in non-small-cell lung cancer. J Clin Oncol 25:587-595, 2007.39. Pirker R, Pereira JR, Szczesna A, et al: Cetuximab plus chemotherapy in patients with advancednon-small-cell lung cancer (FLEX): An open-label randomised phase III trial. Lancet 373:1525-1531,2009.40. Mack PC, Holland WS, Redman M, et al: KRAS mutation analysis in cetuximab-treated advancedstage non-small cell lung cancer (NSCLC): SWOG experience with S0342 and S0536 (abstract 8022).J Clin Oncol 27(15S):412s, 2009.41. O’Byrne KJ, Bondarenko I, Barrios C, et al: Molecular and clinical predictors of outcome forcetuximab in non-small cell lung cancer (NSCLC): Data from the FLEX study (abstract 8007). J ClinOncol 27(15S):408s, 2009.42. Shaw AT, Yeap BY, Mino-Kenudson M, et al: Clinical features and outcome of patients withnon-small-cell lung cancer who harbor EML4-ALK. J Clin Oncol 27:4247-4253, 2009.43. Sandler A, Gray R, Perry MC, et al: Paclitaxel-carboplatin alone or with bevacizumab fornon-small-cell lung cancer. N Engl J Med 355:2542-2550, 2006.44. Reck M, von Pawel J, Zatloukal P, et al: Phase III trial of cisplatin plus gemcitabine with eitherplacebo or bevacizumab as first-line therapy for nonsquamous non-small-cell lung cancer: AVAil. JClin Oncol 27:1227-1234, 2009.45. Patel JD, Hensing TA, Rademaker A, et al: Phase II study of pemetrexed and carboplatin plusbevacizumab with maintenance pemetrexed and bevacizumab as first-line therapy for nonsquamousnon-small-cell lung cancer. J Clin Oncol 27:3284-3289, 2009.46. Pavelic J, Krizanac S, Kapitanovic S, et al: The consequences of insulin-like growthfactors/receptors dysfunction in lung cancer. Am J Respir Cell Mol Biol 32:65-71, 2005.47. Karpp DD, Paz-Ares LG, Novello S, et al: Phase II study of the anti-insulin-like growth factor type 1receptor antibody CP-751,871 in combination with paclitaxel and carboplatin in previously untreated,locally advanced, or metastatic non-small-cell lung cancer. J Clin Oncol 27:2516-2522, 2009.48. Raso MG, Behrens C, Herynk MH, et al: Immunohistochemical expression of estrogen and

Page 9 of 10

Histologic Subtype in NSCLC: Does It Matter?Published on Physicians Practice (http://www.physicianspractice.com)

progesterone receptors identifies a subset of NSCLCs and correlates with EGFR mutation. Clin CancerRes 15:5359-5368, 2009. Source URL: http://www.physicianspractice.com/articles/histologic-subtype-nsclc-does-it-matter

Links:[1] http://www.physicianspractice.com/authors/giovanni-selvaggi-md[2] http://www.physicianspractice.com/authors/giorgio-v-scagliotti-md-phd

Page 10 of 10