《外科护理学》课程阅读资料（英文）肺癌 lung cancer

CA CANCER I CLIN 201161:91-112 Lung Cancer:New Biological Insights and Recent Therapeutic Advances Suresh S.Ramalingam,MD';Taofeek K.Owonikoko,MD,PhD Fadlo R.Khuri,MD Abstract Approximately 1.6 million new cases of lung cancer are diagnosed each vear throughout the world.In many coun cure rates in both re unresectable patient groups.For swith advanced stage disease,modes but real improve and qu t of en a ieved markers for patient selection.Patients with non-small cell lung cancer with mutations in the epidermal growth fac to receptor(EGFR)tyrosine kinase domain achieve response rates of greater than 70%and superior progression-free survival when treated with an EGFR tyrosine kinase inhibitor compared with standard chemotherapy.This has now subset or patients with a mutation in exons e use of an anapla hoa4asea22 Finally,a paradigm shift in favor of maintenance therapy for patients with advanced stage disease has gained strength from recent data.All of these advances have been made possible by developing a greater understanding of the biology,the discovery of novel anticancer agents,and improved supportive care measures.This article reviews ne major strides made in the treatment of lung cancer in the recent past.CA Cancer J Clin 2011;61:91-112. 2011An nerican Cancer Society,Inc Introduction Lnging caue of emortalityrwide with death the year,apprc imately 0aredi 2 There the en ough omen this trend has only been noted very recently in the Unite tes.I ncontrast,in many parts of the worl the number of cases and deaths related to lung cancer is on the rise.' tis also increasingly becoming a disease c the elderly,with a median age at diagnosis of approximately 70 years.Lung cancer is diagnosed at an advanced stage in a majority of patients,which is the primary reason behind the high mortality rate associated with this disease.Early detection continues to be an elusive goal,and substantial numbers of patients diagnosed with local- ized disease are often unsuitable for curative surgical procedures due to concomitant medical illness. Non-small cell lung cancer (NSCLC),which includes adenocarcinoma,squamous cell carcinoma,large cell carcinoma,and bronchioloalveolar carcinoma,accounts for nearly 85%of all cases of lung cancer.Cigarette smoking is the most common etiological factor,accounting for nearly 85%of patients with lung cancer. Associate Professor.De and The Winship Cancer Institute.Emory Unive y and ueta Dis n .epartment of Hem atology and Medical Oncology.Emory Univ ersity School of Medicine.1365 Clifton Road NE 11American Cancer Society,Inc.doi:10.332/c0102. Available online at htto://cajournalong and http://cacancerjoumalorg OLUME 61 NUMBER 2 MARCH/APRIL 201 91

Lung Cancer: New Biological Insights and Recent Therapeutic Advances Suresh S. Ramalingam, MD1 ; Taofeek K. Owonikoko, MD, PhD2 ; Fadlo R. Khuri, MD3 Abstract Approximately 1.6 million new cases of lung cancer are diagnosed each year throughout the world. In many coun￾tries, the mortality related to lung cancer continues to rise. The outcomes for patients with all stages of lung cancer have improved in recent years. The use of systemic therapy in conjunction with local therapy has led to improved cure rates in both resectable and unresectable patient groups. For patients with advanced stage disease, modest but real improvements in overall survival and quality of life have been achieved with systemic chemotherapy. A major focus of research has been the development of molecularly targeted agents and the identification of bio￾markers for patient selection. Patients with non-small cell lung cancer with mutations in the epidermal growth factor receptor (EGFR) tyrosine kinase domain achieve response rates of greater than 70% and superior progression-free survival when treated with an EGFR tyrosine kinase inhibitor compared with standard chemotherapy. This has now emerged as the preferred therapeutic approach for the subset of patients with a mutation in exons 19 or 21 of the EGFR. Another promising targeted approach involves the use of an anaplastic lymphoma kinase (ALK) inhibitor in patients with a translocation involving the echinoderm microtubule-associated protein-like 4 (EML4) and -ALK genes. Finally, a paradigm shift in favor of maintenance therapy for patients with advanced stage disease has gained strength from recent data. All of these advances have been made possible by developing a greater understanding of the biology, the discovery of novel anticancer agents, and improved supportive care measures. This article reviews the major strides made in the treatment of lung cancer in the recent past. CA Cancer J Clin 2011;61:91–112. VC 2011 American Cancer Society, Inc. Introduction Lung cancer is the leading cause of cancer-related mortality worldwide, with nearly 1.4 million deaths each year. 1 Of the 1.6 million new cases of lung cancer diagnosed each year, approximately 220,000 are diagnosed in the United States.2 There has been an overall decrease in the incidence of lung cancer in men, although in women this trend has only been noted very recently in the United States. In contrast, in many parts of the world the number of cases and deaths related to lung cancer is on the rise.1 It is also increasingly becoming a disease of the elderly, with a median age at diagnosis of approximately 70 years.3 Lung cancer is diagnosed at an advanced stage in a majority of patients, which is the primary reason behind the high mortality rate associated with this disease. Early detection continues to be an elusive goal, and substantial numbers of patients diagnosed with local￾ized disease are often unsuitable for curative surgical procedures due to concomitant medical illness. Non-small cell lung cancer (NSCLC), which includes adenocarcinoma, squamous cell carcinoma, large cell carcinoma, and bronchioloalveolar carcinoma, accounts for nearly 85% of all cases of lung cancer. Cigarette smoking is the most common etiological factor, accounting for nearly 85% of patients with lung cancer. 1 Associate Professor, Department of Hematology and Medical Oncology and The Winship Cancer Institute, Emory University, Atlanta, GA; 2 Assistant Professor, Department of Hematology and Medical Oncology and The Winship Cancer Institute, Emory University, Atlanta, GA; 3 Professor and Chair of Hematology and Medical Oncology, Roberto C. Goizueta Distinguished Chair in Translational Cancer Research, Department of Hematology and Medical Oncology and The Winship Cancer Institute, Emory University, Atlanta, GA. Corresponding author: Fadlo R. Khuri, MD, Department of Hematology and Medical Oncology, Emory University School of Medicine, 1365 Clifton Road NE, Atlanta, GA 30322; fkhuri@emory.edu DISCLOSURES: Supported by National Cancer Institute grant 1PO1 CA116676 and the Distinguished Cancer Clinical Scholar Award by the Georgia Cancer Coalition to all the coauthors. Dr. Ramalingam has served on Ad Hoc Advisory Board for Lilly, Genentech, OSI Pharmaceuticals, Astellas, GlaxoSmithKline, and ImClone. Dr. Khuri serves as a consultant for Pfizer. VC 2011 American Cancer Society, Inc. doi:10.3322/caac.20102. Available online at http://cajournal.org and http://cacancerjournal.org VOLUME 61 _ NUMBER 2 _ MARCH/APRIL 2011 91 CA CANCER J CLIN 2011;61:91–112

Lung Cancer Revie Nonetheless,the number of cases of lung ca agn sed in nokers has recent time nd of Mib diseas presence or ce o 1 tion in sm population,I extrathoracic lignant pleural effusion that lung cancer will continue to be a has been upstaged to Mla based on its poorer prog major health problem for the next 40 to 50 years. nosis compared with the rest of the T4 descriptors Small cell lung cancer(SCLC)incidence relative to from the older system.The other major change NSCLC has decreased in recent years,although it involves the categorization of additional nodules in continues to be a lethal disease for the majority of the same lobe and other ipsilateral lobes as T3 and patients. T4,respectively.Although the overall treatment In the past few decades,lung cancer has gone guidelines are not likely to be affected,this new stage from an untreatable disease with dismal outcomes to categorization will lead to improved prognostic one that has a number of novel therapeutic options determination and better stratification of patients on and more recently,to a disease for which individual- tandomized clinical trials. ized treatment options have become available.There has been a tre ndous in Biology of Lung Cancer targets.In d the ionaldinilc be made ba tech lung cance ed on bio earch ha mor,at least for a subset a molecularly diverse set of tumor types whoseon There is a greater emphasis on identifying dominan commonality is their origination in the lung. molecular pathways that drive an individual patient's Lung cancer classification is far more complex thar tumor.This has paved the way for the evaluation of the simplistic grouping into small cell and non-small novel strategies to modulate such molecular targets cell variants that was once thought to represent a ho for the treatment of NSCLC.In addition to molec- ularly targeted approaches,newer chemotherapeuti ume when aiom agents with a favorable therapeutic index have also well accepted that the histologic subdivision of lung been developed.This has enhanced the ability to ancer based on light micros opy uses only one of make refinements to well-established systemic ther many phenotypic manifestatio s of the e methods to char ge control.This article eviews the curately nts in the t nt of appearance oted with the clinically that h ave contribu ted to improved pat ng can ent outcome f a pu to an in ive lung ade alignan Staging of Lung Cancer a nchymal Under the epithelial-m a hallmark of national Association for the Study of Lung Cancer More important to the understanding and manage (IASLC),an updated staging manual for lung cance ment of lung cancer,however,is the crystallization (American Joint Committee on Cancer,seventh edi of the theory that,similar to other cancer types,lung tion)was published and adopted for general use in cancer development is a result of a stepwise progres early 2010.The new staging system incorporates sion of malignant transformation of normal respira- major changes that allow for an improved ability to tory epithelium.5 This transformation is driven by determine overall prognosis.The "T status of the cumulative effect of genetic alterations induced tumors is now categorized into less than 2 cm,2 cm to 3 cm,3 cm to 5 cm,5 cm to 7 cm,and greater smoke,an implicated etiologic factor in more than than 7 cm.The lymph node descriptors remain 85%of all cases of lung cancer diagnosed in the 92

Nonetheless, the number of cases of lung cancer diagnosed in never-smokers has also increased in recent times.4 Despite the positive trend of a reduc￾tion in smoking habits in the general population, it is estimated that lung cancer will continue to be a major health problem for the next 40 to 50 years. Small cell lung cancer (SCLC) incidence relative to NSCLC has decreased in recent years, although it continues to be a lethal disease for the majority of patients.5 In the past few decades, lung cancer has gone from an untreatable disease with dismal outcomes to one that has a number of novel therapeutic options, and more recently, to a disease for which individual￾ized treatment options have become available. There has been a tremendous increase in the understanding of the biological mechanisms that underlie lung can￾cer development, which has led to the identification of novel therapeutic targets. In addition to conven￾tional clinical factors, treatment decisions can now be made based on biological factors related to the tu￾mor, at least for a subset of patients with NSCLC. There is a greater emphasis on identifying dominant molecular pathways that drive an individual patient’s tumor. This has paved the way for the evaluation of novel strategies to modulate such molecular targets for the treatment of NSCLC. In addition to molec￾ularly targeted approaches, newer chemotherapeutic agents with a favorable therapeutic index have also been developed. This has enhanced the ability to make refinements to well-established systemic ther￾apy strategies and to develop newer methods to achieve disease control. This article reviews the recent developments in the treatment of lung cancer that have contributed to improved patient outcomes. Staging of Lung Cancer Under the leadership and coordination of the Inter￾national Association for the Study of Lung Cancer (IASLC), an updated staging manual for lung cancer (American Joint Committee on Cancer, seventh edi￾tion) was published and adopted for general use in early 2010. The new staging system incorporates major changes that allow for an improved ability to determine overall prognosis.6 The ‘‘T’’ status of tumors is now categorized into less than 2 cm, 2 cm to 3 cm, 3 cm to 5 cm, 5 cm to 7 cm, and greater than 7 cm.6 The lymph node descriptors remain unchanged in the new staging system.7 Metastatic disease has been further divided into M1a disease or M1b disease based on the presence or absence of extrathoracic disease.8 Malignant pleural effusion has been upstaged to M1a based on its poorer prog￾nosis compared with the rest of the T4 descriptors from the older system. The other major change involves the categorization of additional nodules in the same lobe and other ipsilateral lobes as T3 and T4, respectively. Although the overall treatment guidelines are not likely to be affected, this new stage categorization will lead to improved prognostic determination and better stratification of patients on randomized clinical trials. Biology of Lung Cancer Advanced molecular biology techniques have greatly accelerated the understanding of cancer biology. The application of such technology to lung cancer research has led to the recognition of lung cancer as a molecularly diverse set of tumor types whose only commonality is their origination in the lung.9-11 Lung cancer classification is far more complex than the simplistic grouping into small cell and non-small cell variants that was once thought to represent a ho￾mogenous tumor population with a comparable out￾come when treated in a similar fashion.12 It is now well accepted that the histologic subdivision of lung cancer based on light microscopy uses only one of many phenotypic manifestations of the genetic changes that underlie lung cancer development. His￾tologic appearance may accurately predict biologic behavior as noted with the clinically indolent course of a pure bronchioloalveolar-type carcinoma relative to an invasive lung adenocarcinoma or the malignant course of a sarcomatoid variant of NSCLC, now known to represent epithelial-mesenchymal transi￾tion, a hallmark of cancer dedifferentiation.13,14 More important to the understanding and manage￾ment of lung cancer, however, is the crystallization of the theory that, similar to other cancer types, lung cancer development is a result of a stepwise progres￾sion of malignant transformation of normal respira￾tory epithelium.15 This transformation is driven by the cumulative effect of genetic alterations induced predominantly by inhaled carcinogens from tobacco smoke, an implicated etiologic factor in more than 85% of all cases of lung cancer diagnosed in the Lung Cancer Review 92 CA: A Cancer Journal for Clinicians

CA CANCER J CLIN 2011:61:91-112 Western world.16.17 The Noguchi classification of (FHIT),epidermal growth factor receptor (EGFR), lung adenocarcinoma is a pioneering effort to relate vclin-der ndent kinase 2a (CDKN2),LKBI,reti- noblastoma (RB),and Mw g mishaps such as chr om of atypical aden hor equer subtypes noma thre ough a progressive transformation into the tages More recently,inversion translocation of type A,B,and C adenocarcinomas with lepidic the echinoderm microtubule-associated protein-like4 growth (referring to growth along alveolar struc- (EML4)and anaplastic lymphoma kinase (LK)genes tures)characterized by an increasing component of on chromosome 2(2p21 and 2p23)was shown to char invasive carcinoma but showing excellent surviva cinealanholbgcPpoiC"7hheanaitng acterize a small subset of NS( outcome,and the type D,E,and F solid-type The discovery of adenocarcinomas with a well-recognized poor prog other molecularly defined lung cancer subsets is likely to nosis.1 Well-designed molecular biology investi be hastened by this finding. gations over the last decade have also led to the Epigenetic modulation leading to changes in the recognition that genetic and molecular alterations in level of DNA methylation or histone acetylation the host as well as in the lung cancer cells drive the The role of biology of These t im netic change in b an t footprints and P can be g prognosis pons dic p16(TP16 ng car ce. for gets of the in refor e rly mole signatures of the dis Pro nyper hylation of between several prognostic markers that have been herin 13(CDH13),TP16,RASSFIA,and APC i patients with early stage NSCLC correlated with arly disease recurrence. Given the regulatory role been described in lung cancer.Genome-Wide Associ of chromatin structure in gene expression and the ation studies revealed a strong association between implication of chromatin structural alteration in can lung cancer risk and a single-nucleotide poly orphism cer development and progression,markers of epige variant in the region of chromosome 15925.1,which netic changes provide early warning signs of cancer contains the nicotinic acetylcholine receptor genes and may also predict the benefit of ep genetic ther (CHRNA5 CHRNA3 and CHrnrA thereby The increasi g number of viding a genetic basis for variation in susceptibility to the addic oducts the for eristic of lung cancer hasno ren into a fo mult Fi .These es will h that on the develc ovative treatmen s for one The NexGer M3 [GSTM3],glutathione ta inf [GSTP1],and glutathione S-transferase theta- [GSTT1D),may predispose toward lung cancer devel gration of genomic information into cancer manage- effect of tobacco carcinogens. The most frequently described acquired genetic (TCGA)is a gargantuan collaborative project built aberrations within the tumor involve the tumor pro- on these technological advancements.It is sponsored tein p53 (TP53),KRAS,fragile histidine triad by the National Cancer Institute (NCI)and the 93

Western world.16,17 The Noguchi classification of lung adenocarcinoma is a pioneering effort to relate tumor histology with clinical and radiologic charac￾teristics.12,15 This has resulted in the identification of atypical adenomatous hyperplasia and adenocarci￾noma in situ as preinvasive neoplastic lung lesions that serve as precursors to invasive lung adenocarci￾noma through a progressive transformation into the type A, B, and C adenocarcinomas with lepidic growth (referring to growth along alveolar struc￾tures) characterized by an increasing component of invasive carcinoma but showing excellent survival outcome, and the type D, E, and F solid-type adenocarcinomas with a well-recognized poor prog￾nosis.12,15 Well-designed molecular biology investi￾gations over the last decade have also led to the recognition that genetic and molecular alterations in the host as well as in the lung cancer cells drive the biology of the tumor. These changes represent im￾portant footprints of biologic drivers and can be used for tumor classification,9-11,18-20 response predic￾tion,19 and assigning prognosis,18,21-23 or even as tar￾gets of therapy. Molecular prognostic models of lung cancer have been developed that are based on genomic signatures of the disease,18-22 or based on the interplay between several prognostic markers that have been determined to be individually important in progno￾sis.24 A very diverse range of genetic abnormalities has been described in lung cancer. Genome-Wide Associ￾ation Studies revealed a strong association between lung cancer risk and a single-nucleotide polymorphism variant in the region of chromosome 15q25.1, which contains the nicotinic acetylcholine receptor genes (CHRNA5, CHRNA3, and CHRNB4), thereby pro￾viding a genetic basis for variation in susceptibility to the addictive potential of tobacco products, the fore￾most etiology of lung cancer.25-28 Further support for an important role for host biology in lung cancer came from the observation that polymorphisms in genes encoding detoxifying enzymes, glutathione S-transferases (glutathione S-transferase Mu 1 [GSTM1], glutathione S-transferase M3 [GSTM3], glutathione S-transferase P [GSTP1], and glutathione S-transferase theta-1 [GSTT1]), may predispose toward lung cancer devel￾opment because of an increased susceptibility to the effect of tobacco carcinogens.29-31 The most frequently described acquired genetic aberrations within the tumor involve the tumor pro￾tein p53 (TP53), KRAS, fragile histidine triad (FHIT), epidermal growth factor receptor (EGFR), cyclin-dependent kinase 2a (CDKN2), LKB1, reti￾noblastoma (RB), and Myc genes.32-34 Larger genomic mishaps such as chromosomal deletions involving the short arms of chromosomes 1, 3, and 9 (del 1p36, del 3p, and del 9p, respectively) are also frequently observed in different lung cancer histologic subtypes and stages.35-38 More recently, inversion translocation of the echinoderm microtubule-associated protein-like 4 (EML4) and anaplastic lymphoma kinase (ALK) genes on chromosome 2 (2p21 and 2p23) was shown to char￾acterize a small subset of NSCLC with a characteristic clinical and histologic profile.39-41 The discovery of other molecularly defined lung cancer subsets is likely to be hastened by this finding. Epigenetic modulation leading to changes in the level of DNA methylation or histone acetylation may result in aberrant gene function. The role of epigenetic changes in lung cancer biology is under￾scored by the finding that aberrant tumor protein p16 (TP16) gene promoter methylation in sputum precedes the detection of lung cancer for up to 3 years in smokers and may therefore represent an early molecular event in the biology of this disease.42 Furthermore, promoter hypermethylation of cad￾herin 13 (CDH13), TP16, RASSFIA, and APC in patients with early stage NSCLC correlated with early disease recurrence.43 Given the regulatory role of chromatin structure in gene expression and the implication of chromatin structural alteration in can￾cer development and progression, markers of epige￾netic changes provide early warning signs of cancer and may also predict the benefit of epigenetic ther￾apy in lung cancer.44-47 The increasing number of molecular, genetic, and histologic alterations charac￾teristic of lung cancer has now been woven into a multistep carcinogenesis model for this cancer type (Fig. 1).15 These biologic advances will have a major impact on the development of innovative treatment options for lung cancer in the years to come. The development of high-throughput NexGen sequencing capability and advanced data information management systems have further advanced the inte￾gration of genomic information into cancer manage￾ment from a mere promise to a realistic possibility in the very near future.48 The Cancer Genome Atlas (TCGA) is a gargantuan collaborative project built on these technological advancements. It is sponsored by the National Cancer Institute (NCI) and the CA CANCER J CLIN 2011;61:91–112 VOLUME 61 _ NUMBER 2 _ MARCH/APRIL 2011 93

EGFR (ype C) 53%) (yP) D21L0H60 13a13L0H53% 7e1879 FIGURE 1.scher tation of Mok o of No r29ReC Cancer National Human Genome Research Institute of the patients with stage I,II,and certain subsets of stage National Institutes of Health to establish genomic IIIA NSCLC.Among patients with early stage changes in the 20 most common cancer types with NSCLC,approximately 40%are not candidates for the goal of facilitating the understanding and ability to urgery due to the presence of medical comorbid illness exploit genomic changes in cancer for dia Removal of an entire lobe of the lung or,in some instances,the entire lung is the recommended approach with ovarian cancer and glioblastoma with plans to for patients with surgically resectable disease.Sublobar cancer ections are associated with a higher incidence Another high-impact res ch endeavor is the Lung nce.In recent tudies have bee Cancer Mutation which ioeloeu uld focused screer ng for driv A .1 d e su on i morbid lung cancer TCGA,it has set specific goals achievabl e in the shor Until the earlier part of this decade,surgery alone to medium term leading to a better understanding of remained the standard of care for patients with early the complex biology of lung cancer and the identifica- stage NSCLC.The use of postoperative radiotherapy tion of predictive,prognostic markers as well as poten- was noted to be detrimental to survival in a meta tial new treatment targets. analysishaled studies cducted in the past 3 decades The included time period included the Early Stage NSCLC era in which older radiation techniques and larger radiation ports were in common use in the absence CuC of prospective studies to address the role of radiation in the postoperative setting,it is not recommended resection is considered the standard therapy for the routine care of patients who underwent a

National Human Genome Research Institute of the National Institutes of Health to establish genomic changes in the 20 most common cancer types with the goal of facilitating the understanding and ability to exploit genomic changes in cancer for diagnostic, prog￾nostic, and therapeutic endpoints. The TCGA began with ovarian cancer and glioblastoma with plans to expand to other tumor types including lung cancer. Another high-impact research endeavor is the Lung Cancer Mutation Consortium study, which was designed to perform focused screening for driver onco￾genic mutations in 1000 patients with stage IIIB/IV NSCLC (ClinicalTrials.gov identifier NCT01014286). Although this is a much more modest effort than the TCGA, it has set specific goals achievable in the short to medium term leading to a better understanding of the complex biology of lung cancer and the identifica￾tion of predictive, prognostic markers as well as poten￾tial new treatment targets. Early Stage NSCLC Approximately one-third of the patients with NSCLC present with early stage disease. Surgical resection is considered the standard therapy for patients with stage I, II, and certain subsets of stage IIIA NSCLC. Among patients with early stage NSCLC, approximately 40% are not candidates for surgery due to the presence of medical comorbid illness. Removal of an entire lobe of the lung or, in some instances, the entire lung is the recommended approach for patients with surgically resectable disease. Sublobar resections are associated with a higher incidence of local recurrence. In recent years, studies have been undertaken to define the subset of patients that could be optimally treated with sublobar resection. Advances in surgical techniques including minimally invasive sur￾gery have led to a reduction in the morbidity and mor￾tality associated with surgery for lung cancer.49-51 Until the earlier part of this decade, surgery alone remained the standard of care for patients with early stage NSCLC. The use of postoperative radiotherapy was noted to be detrimental to survival in a meta￾analysis that analyzed studies conducted in the past 3 decades.52,53 The included time period included the era in which older radiation techniques and larger radiation ports were in common use. In the absence of prospective studies to address the role of radiation in the postoperative setting, it is not recommended for the routine care of patients who underwent a FIGURE 1. Schematic Representation of Molecular Aberrations Implicated in the Multistage Neoplastic Progression of Non-Small Cell Lung Cancer. EGFR indicates epidermal growth factor receptor; LOH, loss of heterozygosity. Reprinted with kind permission from Springer Science 1 Business Media from Noguchi M. Stepwise progression of pulmonary adenocarcinoma–clinical and molecular implications. Cancer Metastasis Rev. 2010;29:15-21. Lung Cancer Review 94 CA: A Cancer Journal for Clinicians

CA CANCER J CLIN 2011:61:91-112 TABLE 1.Randomized trials of Adiuvant chemotherapy for Early stage NSCLc STUDY REGIMEN MEDIAN OR PFS MEDIAN OR S-YEAR OS Adjuvant Lung Project Italy 7839m0佩08995% The Big Lung Trial (BLT) 册 cioo2tug 识6吸9赋9% 5%00g26 a060gP08195% 95320302R26%4 CALGB 96335 pd 盟08说P民8%90% 8品e00 aeoreom sandnmorebne 65700339g00 egafur-uradil vs observation 77.4%s73.6%(P=.25) 8a8%牌05% PFS.progre ee survival:OS. erall survival:HR.hazard ratio:95%Cl,95%confidence interval:CALGB,Car complete surgery.Nonetheless,subset analysis sug in addition to local therapy,the need for ed a ative radiother he rasis has b with fo initial effor with py a high rate of local and systemic failure not su are treated with postoperative systemic Th rapy and radiotherapy. e treatment option for a For patients with early stage NSCLC who are not NSCLC led to studies in patients with early stage candidates for surgery due to medical comorbid ill disease. A meta-analysis reported in 1995 demon ness,palliative external beam radiation is used for strated a modest 14%reduction in the risk of death local tumor treatment and control.Recently,stereo tactic body radiotherapy (SBrT)has emerged as a this was not statistically significant. highly promising therapeutic option for these a number of large clinical trials with adequate power patients.a recent phase 2 study for medically unre to detect modest improvements in outcome have been conducted.Tthe international adiuvant lung safety of using SBRT for patients with tumorsp mal to the c ently under a cisnlatin-hased 2 tion The ults in with gery or ohe ion (Table 1).56-61 Th ectable NSCLC have absolute the 5-year survival ized study in Eu rope that compar SBRT with 49% ith djuvant stage other phase apy udy tha ncy ablation RF is also use da15% solute imp ment n 5-yea patients with early stage NS CLC wh wit he combination c cispla are not candidates for surgical resection.RFA is used patients with stages IB ndha and II NSCLC. A meta-analysis of all recent trials with adjuvant cisplatin-based chemotherapy demor strated a 5%improvement in overall survival with Systemic Chemotherapy adiuvant chemotherapy.These results have led to Nearly two-thirds of recurrences after surgery for the use of adjuvant chemotherapy as the new stand- early stage disease occur at distant sites.Therefore, ard of care for patients with early stage NSCLC. 95

complete surgery. Nonetheless, subset analysis sug￾gested a potential benefit of postoperative radiotherapy in patients with positive surgical margins and those with mediastinal lymph node-positive NSCLC52 that experience a high rate of local and systemic failure and are therefore treated with postoperative systemic therapy and radiotherapy. For patients with early stage NSCLC who are not candidates for surgery due to medical comorbid ill￾ness, palliative external beam radiation is used for local tumor treatment and control. Recently, stereo￾tactic body radiotherapy (SBRT) has emerged as a highly promising therapeutic option for these patients. A recent phase 2 study for medically unre￾sectable patients demonstrated a 3-year primary tumor control rate of 97% with SBRT.54 The efficacy and safety of using SBRT for patients with tumors proxi￾mal to the central airway is currently under investiga￾tion. The promising results in patients with medically unresectable NSCLC have already led to a random￾ized study in Europe that compares SBRT with surgi￾cal resection in patients with early stage NSCLC. Radiofrequency ablation (RFA) is also used for the treatment of patients with early stage NSCLC who are not candidates for surgical resection. RFA is used for the treatment of smaller peripheral tumors and has been associated with high local control rates.55 Systemic Chemotherapy Nearly two-thirds of recurrences after surgery for early stage disease occur at distant sites. Therefore, in addition to local therapy, the need for systemic therapy to eradicate micrometastasis has been recog￾nized for a long time. Despite this, initial efforts to evaluate chemotherapy in the postoperative setting for patients with early stage NSCLC were not suc￾cessful. The emergence of platinum-based chemo￾therapy as an effective treatment option for advanced NSCLC led to studies in patients with early stage disease. A meta-analysis reported in 1995 demon￾strated a modest 14% reduction in the risk of death with the use of cisplatin-based chemotherapy, but this was not statistically significant.53 Subsequently, a number of large clinical trials with adequate power to detect modest improvements in outcome have been conducted. The International Adjuvant Lung Cancer Trial (IALT) included approximately 1800 patients with stages I, II, or III NSCLC who were randomized after surgery to a cisplatin-based, 2- drug regimen or observation (Table 1).56-61 There was an absolute improvement in the 5-year survival rate of 4% with adjuvant chemotherapy. These results were confirmed in another phase 3 study that demonstrated a 15% absolute improvement in 5-year survival with the combination of cisplatin and vinor￾elbine over observation in patients with stages IB and II NSCLC.58 A meta-analysis of all recent trials with adjuvant cisplatin-based chemotherapy demon￾strated a 5% improvement in overall survival with adjuvant chemotherapy.62 These results have led to the use of adjuvant chemotherapy as the new stand￾ard of care for patients with early stage NSCLC. TABLE 1. Randomized Trials of Adjuvant Chemotherapy for Early Stage NSCLC STUDY REGIMEN MEDIAN OR PFS MEDIAN OR 5-YEAR OS Adjuvant Lung Project Italy Mitomycin, vindesine, and cisplatin vs observation 36.5 mo vs 28.9 mo (HR, 0.89; 95% CI, 0.76-1.03 [P ¼ .128]) 55.2 mo vs 48 mo (HR, 0.96; 95% CI, 0.81-1.13 [P ¼ .589]) The Big Lung Trial (BLT)57 Cisplatin-based doublet vs observation 27.0 mo vs 24.7 mo (HR, 0.97; 95% CI, 0.74–1.26 [P ¼ .81]) 33.9 mo vs 32.6 mo (HR, 1.02; 95% CI, 0.77–1.35 [P ¼ .90]) International Adjuvant Lung Cancer Trial (IALT)56 Cisplatin-based doublet vs observation 39.4% vs 34.3% (HR, 0.83; 95% CI, 0.74-0.94 [P < .003]) 44.5% vs 40.4% (HR, 0.86; 95% CI, 0.76-0.98 [P < .03]) JBR1058 Cisplatin and vinorelbine vs observation 61% vs 49% (HR, 0.60; 95% CI, 0.45-0.79 [P < .001]) 94 mo vs 73 mo (HR, 0.69; 95% CI, 0.52-0.91 [P ¼ .04]) CALGB 963359 Carboplatin and paclitaxel vs observation 89 mo vs 56 mo (HR, 0.80; 90% CI, 0.62-1.02 [P ¼ .065]) 95 mo vs 78 mo (HR, 0.83; 90% CI, 0.64-1.08 [P ¼ .125]) Adjuvant Navelbine International Trialist Association (ANITA) Trial60 Cisplatin and vinorelbine vs observation 36.3 mo vs 20.7 mo (HR, 0.76; 95% CI, 0.64–0.91 [P ¼ .002]) 65.7 mo vs 43.7 mo (HR, 0.80; 95% CI, 0.66–0.96 [P ¼ .017]) Japan Lung Cancer Research Group (JLCRG): adjuvant trial61 Tegafur-uracil vs observation 77.4% vs 73.6% (P ¼ .25) 88% vs 85% (HR, 0.71; 95% CI, 0.52-0.98 [P ¼ .04]) NSCLC indicates non-small cell lung cancer; PFS, progression-free survival; OS, overall survival; HR, hazard ratio; 95% CI, 95% confidence interval; CALGB, Can￾cer and Leukemia Group B. CA CANCER J CLIN 2011;61:91–112 VOLUME 61 _ NUMBER 2 _ MARCH/APRIL 2011 95

Lung Cancer Revie esults of the NSCLC66 Recentl wer ot ere was a highe crapy eithe rapy arm years of 回品 tage nerapy wa study therapy in the IALTstudy and the differences superior in the preoperative setting (90%vs 66%) in overall survival and disease-free survival were no Neoadjuvant chemotherapy was associated with a longer statistically significant.In contrast,the sur nonsignificant trend toward longer disease-free sur 中eo9 vival compared with surgery alone.This study was limited in power and included a high proportion of although many of the adiuvant therapy studies patients with stage I disease,a group in whom the have included patients with stages I,II,and III dis- role of chemotherapy is unproven.These studies ease,among patients with stage I disease.the benefit suggest that neoadjuvant therapy is an efficacious from adjuvant chemotherapy has not been clearly and safe approach for patients with early stage established.In fact,the me -analysis noted a detri NSCLC.The clinical circumstances under which it mental hazard ratio for patients with stage IA dis may be preferred over adjuvant thera treated with othe dard of y defined and ng p tthgeB ng pr studies nodes iciar n to cho efit fo 59 otfcad regimens is unpro the effi therapy ir crapy in patic re fo on cted by the Can- ntegration of rly targeted agents and indi cer and Leukemia Group B(CALGB),despite an vidualization of therapy based on biomarke improvement in disease-free survival,there was no epair cross-complementir g gene 1 (ERCC1)is an mportant mediator of DNA repair and has been noted rative intent of therapy in this setting,the small paunum hem the In a subset analysis of the advantage with cisplatin over carboplatin might be an IALT,tumor specimens of patients were evaluated by important reason to use the former.The optimal num mmunohistochemistr for ERCC1 expression. Patients with high ERCC1 expression did not appear d on the pre sen to derive benefit from cisplatin-based chemother evidence,3 to 4 cvcles of cisplatin-based chem otherapy py h the overall survival for this was m are administered in routin In all of the studiesh ble In the ERCC1-ne evaluat ciated with chemotherapy,deliver of plan n This led 60% to the es that eval te t ERCC1 expression for patie nts with early surgery fo tage NSC lung cancer. To improve the delivery of The EGFR inhibitors have also been investigated therapy in patients with early stage NSCLC,the therapy for patients with early stage neoadjuvant approach has been investigated exten NSCLC.In a recent report,the use of gefitinib was sively.A phase 3 study conducted by the Southwes Oncology Group demonstrated an improvement in tebo in patients with resected early stage NSCLC Notably,there was no benefit noted with gefitinib ver.these differen even in the small percentage of patients(21%)with an ces did not reach statistical significance because the EGER mutation.Pertinent limitations of this study trial was closed early Similar trends were also noted nclude its early closure before full accrual and the in a European study that evaluated preoperative short median duration of gefitinib therapy.Another chemotherapy for patients with early stage study that evaluated erlotinib in the postoperative 96

Recently, the long-term follow-up results of the IALT and the NCI-Canada’s JBR10 study were reported. There was a higher incidence of noncancer deaths in the chemotherapy arm after 5 years of study therapy in the IALT study and the differences in overall survival and disease-free survival were no longer statistically significant.63 In contrast, the sur￾vival benefit with chemotherapy was maintained at 9 years of follow-up in the JBR10 study.64 Although many of the adjuvant therapy studies have included patients with stages I, II, and III dis￾ease, among patients with stage I disease, the benefit from adjuvant chemotherapy has not been clearly established. In fact, the meta-analysis noted a detri￾mental hazard ratio for patients with stage IA disease who were treated with adjuvant chemotherapy.62 Among patients with stage IB disease, subset analyses of randomized studies have documented a modest benefit for patients with tumors larger than 4 cm.59 The role of carboplatin-based regimens is unpro￾ven as adjuvant therapy in patients with early stage NSCLC. In a phase 3 study conducted by the Can￾cer and Leukemia Group B (CALGB), despite an improvement in disease-free survival, there was no survival benefit reported with carboplatin and pacli￾taxel in patients with stage IB disease.59 Given the cu￾rative intent of therapy in this setting, the small advantage with cisplatin over carboplatin might be an important reason to use the former. The optimal num￾ber of cycles of adjuvant chemotherapy has not been addressed in randomized studies. Based on the present evidence, 3 to 4 cycles of cisplatin-based chemotherapy are administered in routine practice settings. In all of the studies that evaluated postoperative chemotherapy, delivery of planned courses of chem￾otherapy was achieved in only 60% to 75% of the patients. This is likely related to comorbid illness and the varying recovery period after surgery for lung cancer. To improve the delivery of systemic therapy in patients with early stage NSCLC, the neoadjuvant approach has been investigated exten￾sively. A phase 3 study conducted by the Southwest Oncology Group demonstrated an improvement in overall survival with chemotherapy followed by sur￾gery versus surgery alone.65 However, these differen￾ces did not reach statistical significance because the trial was closed early. Similar trends were also noted in a European study that evaluated preoperative chemotherapy for patients with early stage NSCLC.66 Recently, a phase 3 study compared the administration of chemotherapy either before or af￾ter surgery with surgery alone in patients with early stage NSCLC.67 The delivery of chemotherapy was superior in the preoperative setting (90% vs 66%). Neoadjuvant chemotherapy was associated with a nonsignificant trend toward longer disease-free sur￾vival compared with surgery alone. This study was limited in power and included a high proportion of patients with stage I disease, a group in whom the role of chemotherapy is unproven. These studies suggest that neoadjuvant therapy is an efficacious and safe approach for patients with early stage NSCLC. The clinical circumstances under which it may be preferred over adjuvant therapy, the current standard of care, have not yet been clearly defined and it is therefore left to the judgment of the treating phy￾sician to choose the appropriate setting for systemic therapy in patients with early stage NSCLC. Presently, efforts to improve the efficacy of systemic therapy in patients with early stage NSCLC are focused on the integration of molecularly targeted agents and indi￾vidualization of therapy based on biomarkers. Excision repair cross-complementing gene 1 (ERCC1) is an important mediator of DNA repair and has been noted in several studies to be a determinant of sensitivity to platinum-based therapy.68,69 In a subset analysis of the IALT, tumor specimens of patients were evaluated by immunohistochemistry for ERCC1 expression.70 Patients with high ERCC1 expression did not appear to derive benefit from cisplatin-based chemotherapy, although the overall survival for this group was more favorable. In the ERCC1-negative group, adjuvant chemotherapy was associated with a robust survival advantage over observation. This observation has led to prospective studies that evaluate treatment selection based on ERCC1 expression for patients with early stage NSCLC. The EGFR inhibitors have also been investigated as adjuvant therapy for patients with early stage NSCLC. In a recent report, the use of gefitinib was associated with inferior results compared with pla￾cebo in patients with resected early stage NSCLC.71 Notably, there was no benefit noted with gefitinib, even in the small percentage of patients (21%) with an EGFR mutation. Pertinent limitations of this study include its early closure before full accrual and the short median duration of gefitinib therapy. Another study that evaluated erlotinib in the postoperative Lung Cancer Review 96 CA: A Cancer Journal for Clinicians

CA CANCER J CLIN 2011:61:91-112 setting has completed accrual and the results ar of many studies.The current standard of 60 to 66 awaited.Other strategies currently under evaluation for grays (Gy)was established by a randomized study ombination of be ducted in the 1970s.7s With the availability ith chemothe apy genomic approac he mp ved radiothe it has be for risk stratification ible higher do radiotherap city.Se Surgically Unresectable Stage Ill Disease than 70 Gy and noted promis sed on this,a randomized study is currently Locally advanced disease that is not amenable to sur- inderway to compare 74 Gy of radiotherapy with gical resection is treated with combined modality the standard 60-Gy dose for patients with locally approaches involving systemic therapy and radio advanced NSCLO The use of hyperfractionated therapy.Typically,tumors that directly invade the radiotherapy has also been associated with favorable mediastinum.maior blood vessels.heart.or the ver results over standard once-daily fractionation in tebral body are considered surgically unresectable.In randomized studies,although this approach has been addition ratients with multistation n2 disease are also considered candidates for definitive combined modality treatment approaches. ve the benefit of sys mic therapy The addition of sy mic chemother to radio the inte tion of t reted ager ther nt in and the devel Alth EGFR kinase inhib rand The initial adiothe the ng strateg ves th nb n a pha and neck cancer,the addi ton of tuximab to radi tial approach. Although the con current approach is associated with higher toxicity,there is a modest the overall survival. Based on this,cetuximab improvement in overall survival as well.Therefore,for now being tested in combination with chemoradic patients with a good performance status.concurrent therapy for the treatment of patients with locally chemoradiotherapy has become the standard of care advanced nsClC in a phase 3 study (Radiation Two different chemotherapy oaches are com Therapy Oncology Group [RTOG]0617).A monly used for locally advanced NSCLC.One e randomized study that evaluated gefitinib as mainte- men involves the use of full doses of cisplatin and nance therapy after combined modality therapy etoposide with radiotherapy.76 The other demonstrated inferior survival compared with pla cebo.4 The use of antiangiogenic ag ents has under reekly schedule at "radio itiz II dis cvcles of full-dose the conducted in oted th approaches haveen y in the al fistula in pat rren che reated ith acizun etopos e regin ith higher to servation a the of t ha he pote l eding,the dey opment of b therapy stat an Conversely n patients with lod the weekly regime of carboplatin and paclitaxel has o be pursued furthe a more favorable tolerability profile and can be com- The use of induction or consolidation chemother bined with higher doses of radiotherapy.These 2 apy has not been associated with improvement in strategies have not been directly compared in randomized clinical trials. Recently,pemetrexed has been com- The optimal dose of radiotherapy for patients bined with cisplatin or carboplatin in full systemic with locally advanced NSCLC has been the subject doses with concurrent radiotherapy.Promising results 97

setting has completed accrual and the results are awaited. Other strategies currently under evaluation for early stage NSCLC include the combination of bevaci￾zumab with chemotherapy and genomic approaches for risk stratification. Surgically Unresectable Stage III Disease Locally advanced disease that is not amenable to sur￾gical resection is treated with combined modality approaches involving systemic therapy and radio￾therapy. Typically, tumors that directly invade the mediastinum, major blood vessels, heart, or the ver￾tebral body are considered surgically unresectable. In addition, patients with multistation N2 disease are also considered candidates for definitive combined modality treatment approaches. The addition of systemic chemotherapy to radio￾therapy was associated with improvement in overall survival compared with radiotherapy alone in randomized studies.72,73 The initial trials used a sequential approach involving the 2 modalities. Subsequently, a number of studies have compared the concurrent use of chemoradiotherapy with the sequen￾tial approach.74,75 Although the concurrent approach is associated with higher toxicity, there is a modest improvement in overall survival as well. Therefore, for patients with a good performance status, concurrent chemoradiotherapy has become the standard of care. Two different chemotherapy approaches are com￾monly used for locally advanced NSCLC. One regi￾men involves the use of full doses of cisplatin and etoposide with radiotherapy.76 The other approach involves the use of carboplatin and paclitaxel on a weekly schedule at ‘‘radiosensitizing’’ doses, followed by 2 cycles of full-dose therapy.77 Both of these approaches have demonstrated efficacy in the con￾current chemoradiotherapy setting. The cisplatin and etoposide regimen is associated with higher tox￾icity, but has the potential to eradicate micrometa￾static disease at an earlier time point. Conversely, the weekly regimen of carboplatin and paclitaxel has a more favorable tolerability profile and can be com￾bined with higher doses of radiotherapy. These 2 strategies have not been directly compared in randomized clinical trials. The optimal dose of radiotherapy for patients with locally advanced NSCLC has been the subject of many studies. The current standard of 60 to 66 grays (Gy) was established by a randomized study conducted in the 1970s.78 With the availability of improved radiotherapy technology, it has become possible to deliver even higher doses of radiotherapy without a substantial increase in toxicity. Several phase 2 studies have evaluated radiotherapy doses higher than 70 Gy and noted promising results.79,80 Based on this, a randomized study is currently underway to compare 74 Gy of radiotherapy with the standard 60-Gy dose for patients with locally advanced NSCLC. The use of hyperfractionated radiotherapy has also been associated with favorable results over standard once-daily fractionation in randomized studies, although this approach has been limited by the logistical challenges in delivering mul￾tiple fractions of treatment on a daily basis.81,82 Efforts to improve the benefit of systemic therapy have focused on the integration of targeted agents and the development of newer chemotherapeutic agents. Although EGFR tyrosine kinase inhibitors can be safely administered with chemoradiotherapy, the most promising strategy involves the use of cetuximab. In a phase 3 study of patients with head and neck cancer, the addition of cetuximab to radio￾therapy resulted in nearly a 2-fold improvement in the overall survival.83 Based on this, cetuximab is now being tested in combination with chemoradio￾therapy for the treatment of patients with locally advanced NSCLC in a phase 3 study (Radiation Therapy Oncology Group [RTOG] 0617). A randomized study that evaluated gefitinib as mainte￾nance therapy after combined modality therapy demonstrated inferior survival compared with pla￾cebo.84 The use of antiangiogenic agents has under￾gone limited evaluation for stage III disease. Two studies conducted in patients with SCLC noted the development of tracheoesophageal fistula in patients treated with bevacizumab after radiotherapy.85 Because of this observation and the higher risk of bleeding, the development of bevacizumab therapy in patients with locally advanced NSCLC is unlikely to be pursued further. The use of induction or consolidation chemother￾apy has not been associated with improvement in overall survival in patients with locally advanced NSCLC.86,87 Recently, pemetrexed has been com￾bined with cisplatin or carboplatin in full systemic doses with concurrent radiotherapy. Promising results CA CANCER J CLIN 2011;61:91–112 VOLUME 61 _ NUMBER 2 _ MARCH/APRIL 2011 97

noted in a randomized phase 2 study by the ands,both CALGB that of NSCLC I n extens y st gene boplati lity profl the cisplatin and pemetrexed regimen with the cispla- over cisplatin-based regimens. Despite the margin tin and etoposide regimen in combination with radic ally higher response rate noted with cisplatin-based therapy for patients with locally advanced NSCLC regimens,considering the palliative intent of therapy, Multimodality approaches result in cure for carboplatin-based regimens have found wide applic approximately 20%of the patients with locally ability in routine care.However.recent improvements advanced NSCLC.Both improvements in svstemi in antiemetic therapy have rendered the use of therapy and local therapy have contributed to the cisplatin-based regimens more tolerable. more favorable patient outcome noted in recent A pumber of randomized clinical trials have estab vears Another factor that warrants mention is the ishcdhtcsupciotiyofaplhatinum-containing role of positron emission tomography,which has drug combination over single-agent therapy. improved the accuracy of stagi The response rate,progre -free rvival,and As esult,stage m igration of lung cancer. rall survival all a o be 6 mpr to the ments in outcom pat nts tage fits xicity.Pa eta Advanced Stage NSCLC ine. om to as the"thi eration"cytotoxic agents ng cance have advanced stage disease at the time of diagnosis with a platinum with Even patients without any cancer-related symptoms advanced NSCLC A large randomized at diagnosis will manifest symptoms as their disease clinical trial conducted by the Eastern Cooperative progresses.Therefore,the overall goals of treatment Oncology Group (ECOG)compared the efficacy of are to improve symptoms,preserve or improve qual 4 commonly used combination regimens in the ity of life.and prolong survival.The performance treatment of patients with advanced NSCLC.97 The status of the patient remains an important determi- ombinations of cisplatin and docetaxel.cisplatir nant of overall ore nosis and is a nrime considera and gemcitabine,and carboplatin and paclitaxel were tion in treatment selection.Recently,gender has also all associated with efficacy become recognized as an important prognostic fac- tor,with females exp periencing ival thar avel remi en.These obser vations rted b nales on the database for the nev of other taging able stablished acy plateau"had rath met to the of patients with oxicity without ar evide nce of an improvement i me dise sease to Mla and M1b efficacy across clinical trials Platinum-Based Chemotherapy available platinum-based 2-drug regimens,the me Systemic therapy remains the mainstay of treatment dian survival and one-vear survival rate are 8 months of advanced stage nsclc.Combination chemo therapy with a platinum-based regimen has emerged non分nn as standard therapy for patients with advanced st disease.Improvements in overall survival and qualit Role of Histology of life have been demonstrated with platinum-based Until ll histolocal suypes of NSCLC regimens over supportive care alone in randomized were treated alike because a diffe ntial sensitivity 8

were noted in a randomized phase 2 study by the CALGB that evaluated carboplatin and pemetrexed with concurrent radiotherapy.88 These favorable results have prompted an ongoing phase 3 study to compare the cisplatin and pemetrexed regimen with the cispla￾tin and etoposide regimen in combination with radio￾therapy for patients with locally advanced NSCLC.89 Multimodality approaches result in cure for approximately 20% of the patients with locally advanced NSCLC. Both improvements in systemic therapy and local therapy have contributed to the more favorable patient outcome noted in recent years. Another factor that warrants mention is the role of positron emission tomography, which has improved the accuracy of staging of lung cancer.90 As a result, stage migration could also have contrib￾uted to the recent improvements in outcome for patients with stage III disease. Advanced Stage NSCLC Approximately 50% of the patients with lung cancer have advanced stage disease at the time of diagnosis. Even patients without any cancer-related symptoms at diagnosis will manifest symptoms as their disease progresses. Therefore, the overall goals of treatment are to improve symptoms, preserve or improve qual￾ity of life, and prolong survival. The performance status of the patient remains an important determi￾nant of overall prognosis and is a prime considera￾tion in treatment selection. Recently, gender has also become recognized as an important prognostic fac￾tor, with females experiencing a better survival than males.91-94 In addition, the database for the new staging system demonstrated a more favorable out￾come for patients without extrathoracic disease, with a median survival of 14 months compared with only 6 months in those with distant metastasis. This led to the additional subclassification of patients with metastatic disease to M1a and M1b. Platinum-Based Chemotherapy Systemic therapy remains the mainstay of treatment of advanced stage NSCLC. Combination chemo￾therapy with a platinum-based regimen has emerged as standard therapy for patients with advanced stage disease.95 Improvements in overall survival and quality of life have been demonstrated with platinum-based regimens over supportive care alone in randomized clinical trials.96 Among the platinum compounds, both cisplatin and carboplatin have been extensively studied for the treatment of NSCLC. In general, carboplatin￾based regimens have a favorable tolerability profile over cisplatin-based regimens.97,98 Despite the margin￾ally higher response rate noted with cisplatin-based regimens, considering the palliative intent of therapy, carboplatin-based regimens have found wide applic￾ability in routine care. However, recent improvements in antiemetic therapy have rendered the use of cisplatin-based regimens more tolerable. A number of randomized clinical trials have estab￾lished the superiority of a platinum-containing, 2- drug combination over single-agent therapy.99-101 The response rate, progression-free survival, and overall survival all appear to be improved with com￾bination regimens in patients with advanced NSCLC, although the benefits come with higher toxicity. Paclitaxel, docetaxel, gemcitabine, vinorel￾bine, irinotecan, and pemetrexed, commonly referred to as the ‘‘third-generation’’ cytotoxic agents, have all demonstrated efficacy when given in combination with a platinum compound in patients with advanced NSCLC.97,99,102-105 A large randomized clinical trial conducted by the Eastern Cooperative Oncology Group (ECOG) compared the efficacy of 4 commonly used combination regimens in the treatment of patients with advanced NSCLC.97 The combinations of cisplatin and docetaxel, cisplatin and gemcitabine, and carboplatin and paclitaxel were all associated with efficacy parameters comparable to those of the control arm of the cisplatin and pacli￾taxel regimen. These observations, supported by the findings of other contemporaneous clinical trials, established the notion that an ‘‘efficacy plateau’’ had been reached with 2-drug combinations in patients with advanced stage NSCLC. The use of 3-drug cy￾totoxic combinations has generally resulted in higher toxicity without clear evidence of an improvement in efficacy across clinical trials and has therefore not been pursued subsequently.106 With the currently available platinum-based 2-drug regimens, the me￾dian survival and one-year survival rate are 8 months to 11 months and 30% to 40%, respectively, in patients with a good performance status.107 Role of Histology Until recently, all histological subtypes of NSCLC were treated alike because a differential sensitivity Lung Cancer Review 98 CA: A Cancer Journal for Clinicians

CA CANCER J CLIN 2011:61:91-112 based on histology had not been appreciated in clini- differences in sensitivity to certain chemotherap regimens represent an underlying biological facto results of a phase 3 study that co athcrthanmcrehistologicaldiferencesinfumor of cisplatin and p exed with that c with advanced Maintenance Therapy f )patients,thi a pre plan to analyze outcomes based on histology.For the nce the role overall study population,the median survival was Randomized studies that compared the use of combi similar at 10.3 months with both regimens.How nation chemotherapy for a defined number of cycles ever,in patients with adenocarcinoma histology,the versus continuation until progression or a higher cisplatin and pemetrexed regimen was associated number of courses failed to demonstrate an advantage with a superior median survival(12.6 months vs 10.9 for the latter approach.Therefore.the administration months).In addition,this regimen was also associ- of 4 to 6 cycles of combination chemotherapy fol- ated with a favorable tolerability profile These lowed by observation became the standard of care for results led to the approval of the cisplatin and peme- the first-line treatment of advanced NSClc With trexed regimen for patients with nonsqua mous the advent of well-tolerated novel chemotheraneutic NSCLC.thus setting patients with NSC evaluated the f singl ance the The biological reasons behind the highe sen after the achie of ade clea to pe exp This com monly referred to a rapy or on t rapy.In a tre as give anc ade squamous or ediately after cycle ion therapy with carboplatin and gemcitabine or after progre The preliminary results of a phase 3 study tha compared the use of carboplatin in combination with Only patients either nanoparticle albumin-bound (nab)paclitaxel with an objective response or stable disease were or standard paclitaxel formulation as the first-line tandomized to maintenance therapy.There was a therapy for patients with advanced NSCLC also significant improvement in progression-free survival noted a differential effect by histology.11 Response and a trend toward a survival benefit for patients rate,the primary endpoint of the study.was superior reated with maintenance therapy.App ximately for the overall study population (33%vs 25%).and 4006 of the natients on the control arm did no in patients with squamous histology,there receive the planned second-line therapy for a variety vith nab paclitax (419%vs240% ns including dise .For pa ived the ned d-line the py the the ian of S which facilit rec d ma therapy 1 SPARC of the agent incre expression has to a highe anoth active c emotherapy agent response rate in patients with head and neck cancers, mmediately following the frontline therapy before which are predominantly of squamous histology disease progression. The nab paclitaxel regimen was also associated with In another phase 3 study with a different design a lower incidence of neuropathy.The survival data patients with nonprogressive disease after 4 cycles of from this study will be the decisive factor in how this regimen is used in routine practice.Although these thp observations are interesting,it is likely that the the overall patient population,the primary endpoint 99

based on histology had not been appreciated in clini￾cal trials. Recently, Scagliotti et al reported the results of a phase 3 study that compared the combi￾nation of cisplatin and pemetrexed with that of cis￾platin and gemcitabine for patients with advanced stage NSCLC.104 With a sample size of approxi￾mately 1700 patients, this study had a predefined plan to analyze outcomes based on histology. For the overall study population, the median survival was similar at 10.3 months with both regimens. How￾ever, in patients with adenocarcinoma histology, the cisplatin and pemetrexed regimen was associated with a superior median survival (12.6 months vs 10.9 months). In addition, this regimen was also associ￾ated with a favorable tolerability profile. These results led to the approval of the cisplatin and peme￾trexed regimen for patients with nonsquamous NSCLC, thus setting the precedent for histology￾based treatment selection for patients with NSCLC. The biological reasons behind the higher sensitivity of adenocarcinoma to pemetrexed are not entirely clear. Some studies have linked the expression of thymidylate synthase (TS), a known target for peme￾trexed, with efficacy.108,109 TS expression is lower in adenocarcinoma compared with squamous or SCLC.110 The preliminary results of a phase 3 study that compared the use of carboplatin in combination with either nanoparticle albumin-bound (nab) paclitaxel or standard paclitaxel formulation as the first-line therapy for patients with advanced NSCLC also noted a differential effect by histology.111 Response rate, the primary endpoint of the study, was superior for the overall study population (33% vs 25%), and in patients with squamous histology, there was a greater advantage with nab paclitaxel (41% vs 24%). A biological rationale for this observation might be the expression of Secreted Protein Acidic and Rich in Cysteine (SPARC), which facilitates the accumu￾lation of albumin in the tumor and thus increases in￾tracellular concentrations of the cytotoxic agent.112 SPARC expression has been linked to a higher response rate in patients with head and neck cancers, which are predominantly of squamous histology. The nab paclitaxel regimen was also associated with a lower incidence of neuropathy. The survival data from this study will be the decisive factor in how this regimen is used in routine practice. Although these observations are interesting, it is likely that the differences in sensitivity to certain chemotherapy regimens represent an underlying biological factor rather than mere histological differences in tumors. Maintenance Therapy The optimal duration of treatment for patients with advanced NSCLC has been the subject of study ever since the role of chemotherapy was established.113,114 Randomized studies that compared the use of combi￾nation chemotherapy for a defined number of cycles versus continuation until progression or a higher number of courses failed to demonstrate an advantage for the latter approach. Therefore, the administration of 4 to 6 cycles of combination chemotherapy fol￾lowed by observation became the standard of care for the first-line treatment of advanced NSCLC. With the advent of well-tolerated, novel chemotherapeutic and molecularly targeted agents, recent studies have evaluated the role of single-agent maintenance ther￾apy after the achievement of maximal disease control with combination regimens. This strategy is now commonly referred to as maintenance therapy or consolidation therapy. In a phase 3 study, docetaxel was given as maintenance immediately after 4 cycles of combination therapy with carboplatin and gemcitabine or after progres￾sion of disease (second-line therapy) in patients with advanced NSCLC (Table 2).113-123 Only patients with an objective response or stable disease were randomized to maintenance therapy. There was a significant improvement in progression-free survival and a trend toward a survival benefit for patients treated with maintenance therapy. Approximately 40% of the patients on the control arm did not receive the planned second-line therapy for a variety of reasons including disease progression. For patients who received the planned second-line therapy, the survival outcomes were similar to those of patients who received maintenance therapy. This observation suggests that the benefit from maintenance therapy is probably related to the increased likelihood of administration of another active chemotherapy agent immediately following the frontline therapy before disease progression. In another phase 3 study with a different design, patients with nonprogressive disease after 4 cycles of platinum-based combination therapy were random￾ized to therapy with pemetrexed or placebo.124 For the overall patient population, the primary endpoint CA CANCER J CLIN 2011;61:91–112 VOLUME 61 _ NUMBER 2 _ MARCH/APRIL 2011 99

Lung Cancer Revi TABLE 2.Clinical Trials of Maintenance Therapy in Advanced NSCLC STRATEGY CONTINUATION THERAPY WITH ORIGINAL CHEMOTHERAPY AGENT TRIAL REGIMEN PES Ios 5mh200114 mo ys 5 mo 6 mo ys 7 mo 0t2002 85m0566mP=63 Brodowicz 2006 GC vs GC followed by G 5m0s6.6m0P=.001】 11m0s13m0(P=.195) Park 2007 PC x 4 cydes ys PC x 6 ccles 6.2m0543m0P=001) 15mos16m0.(P=.4691 Belani 2010 4m01.7m0P=575) 8m0s9.3m0P=83g) STRATEGY WITCH MAINTENANCE Vesteel 20059 MIC ys MIC followed by vinorelbine 5m0s3m0(P=32) 10.4m0s11mc Fias2007120 G and carboplatin follow 7mos27m0(P=001) 23m0sg7m0P=853 Cappuzzo (SATURN trial)2010 Platinum doublet followed by placebo vs erlotini 11.1k123k(P=.001 11m0s12m0P=.0088 Miller (ATLAS trian 2009z2 可 476m053.75m0(P■0012) Not reported Perol (IFCT-GFPC 0502)2010 3.8 mo ys 1.9 mo for G (P =.001) of median progression-free survival was superior this difference did not translate into a survival bene with pemetrexed therapy.The overall survival was fit.This is likely due to the fact that the majority of also superior with maintenance therapy.As noted in patients with an EGFR mutation on the placebo arm the frontline setting.the benefit from maintenance crossed over to receive an eger inhibitor after dis therapy was restricted to patients with nonsquamous ease progression.For patients with wild-type EGFR histology.Only 67%of the patients on the control a modest improvement in overall survival was noted arm rec eived second-line therany which could have with erlotinib.Approximately 70%of patients in contributed to the favorable outcome with mainte hoth arms received subsequent lines of therapy.The FDa has recently apr ed erlotinib for ival by 5 months with ed in y for with advanced NSCLC t s histology led to its The favo with and Drg dn inis antiated by a-analysi (FDA)for main utilize m intenanc therapy It no Impro ent in ove rall survival has also been ment in progression- improe a m noted with the use of erlotinib,an EGFR inhibito survival with the mainte as maintenance therapy.In a ph s study reported nance strategy. recently,patients with advanced NSCLC with dis- Several important questions about the optimal utilization of maintenance therapy still remain and will be addressed by ongoing/planned clinical trials. There was a modest improvement in the median One question relates to the use of an agent used in progression-free survival(12.3 weeks vs 11.1 weeks) the first-line combination versus switching to an and overall survival (12 months vs 11 months)in alternate agent for maintenance therapy Unti favor of erlotinib.The presence of activating muta recently,all the studies that showed a stro ng benefit tions in the EGFR was highly predictive of benefit. with maintenance therapy used an alternative agent with a hazard ratio of 0.10 with erlotinib,although In a recent phase 3 study,after frontline therapy 100

of median progression-free survival was superior with pemetrexed therapy. The overall survival was also superior with maintenance therapy. As noted in the frontline setting, the benefit from maintenance therapy was restricted to patients with nonsquamous histology. Only 67% of the patients on the control arm received second-line therapy, which could have contributed to the favorable outcome with mainte￾nance therapy. The robust overall improvement in Median Survival by 5 months with pemetrexed in patients with nonsquamous histology led to its ap￾proval by the US Food and Drug Administration (FDA) for maintenance therapy for patients with nonsquamous NSCLC. Improvement in overall survival has also been noted with the use of erlotinib, an EGFR inhibitor, as maintenance therapy. In a phase 3 study reported recently, patients with advanced NSCLC with dis￾ease control after 4 cycles of platinum-based therapy were randomized to receive erlotinib or placebo.121 There was a modest improvement in the median progression-free survival (12.3 weeks vs 11.1 weeks) and overall survival (12 months vs 11 months) in favor of erlotinib. The presence of activating muta￾tions in the EGFR was highly predictive of benefit, with a hazard ratio of 0.10 with erlotinib, although this difference did not translate into a survival bene- fit. This is likely due to the fact that the majority of patients with an EGFR mutation on the placebo arm crossed over to receive an EGFR inhibitor after dis￾ease progression. For patients with wild-type EGFR, a modest improvement in overall survival was noted with erlotinib. Approximately 70% of patients in both arms received subsequent lines of therapy. The FDA has recently approved erlotinib for mainte￾nance therapy for patients with advanced NSCLC based on this study. The favorable outcome with these trials was also substantiated by a meta-analysis of all studies that utilized some form of maintenance therapy. It noted a statistically significant improve￾ment in progression-free survival and a modest improvement in overall survival with the mainte￾nance strategy.125 Several important questions about the optimal utilization of maintenance therapy still remain and will be addressed by ongoing/planned clinical trials. One question relates to the use of an agent used in the first-line combination versus switching to an alternate agent for maintenance therapy. Until recently, all the studies that showed a strong benefit with maintenance therapy used an alternative agent. In a recent phase 3 study, after frontline therapy TABLE 2. Clinical Trials of Maintenance Therapy in Advanced NSCLC STRATEGY CONTINUATION THERAPY WITH ORIGINAL CHEMOTHERAPY AGENT TRIAL REGIMEN PFS OS Smith 2001114 MVP
3 cycles vs MVP
6 cycles 5 mo vs 5 mo 6 mo vs 7 mo Socinski 2002113 PC
4 cycles vs PC until progression — 8.5 mo vs 6.6 mo (P ¼ .63) Brodowicz 2006116 GC vs GC followed by G 5 mo vs 6.6 mo (P ¼ .001) 11 mo vs 13 mo (P ¼ .195) Park 2007117 PC
4 cycles vs PC
6 cycles 6.2 mo vs 4.3 mo (P ¼ .001) 15 mo vs 16 mo (P ¼ .469) Belani 2010118 G and carboplatin followed by BSC vs G and carboplatin followed by G 7.4 mo vs 7.7 mo (P ¼ .575) 8 mo vs 9.3 mo (P ¼ .838) STRATEGY SWITCH MAINTENANCE Westeel 2005119 MIC vs MIC followed by vinorelbine 5 mo vs 3 mo (P ¼ .32) 10.4 mo vs 11 mo Fidias 2007120 G and carboplatin followed immediately by D vs G and carboplatin followed by D at progression 5.7 mo vs 2.7 mo (P ¼ .001) 12.3 mo vs 9.7 mo (P ¼ .853) Cappuzzo (SATURN trial) 2010121 Platinum doublet followed by placebo vs erlotinib 11.1 wk vs 12.3 wk (P ¼ .001) 11 mo vs 12 mo (P ¼ .0088) Miller (ATLAS trial) 2009122 Platinum doublet plus bevacizumab followed by bevacizumab vs erlotinib plus bevacizumab 4.76 mo vs 3.75 mo (P ¼ .0012) Not reported Perol (IFCT-GFPC 0502) 2010123 GC followed by BSC vs G vs erlotinib 2.9 mo vs 1.9 mo for erlotinib (P ¼ .002) 3.8 mo vs 1.9 mo for G (P ¼ .001) NSCLC indicates non-small cell lung cancer; PFS, progression-free survival; OS, overall survival; MVP, mitomycin, vinblastine, and cisplatin; PC, paclitaxel and carboplatin; GC, gemcitabine and cisplatin; BSC, best supportive care; MIC, mitomycin, ifosfamide, and cisplatin; D, docetaxel; SATURN, Sequential Tarceva in Unresectable NSCLC. Lung Cancer Review 100 CA: A Cancer Journal for Clinicians