Disciplined Practice and Improving Clinical and Pathologic Staging for Non-Small Cell Lung Cancer
3
Disciplined Practice and Improving Clinical and Pathologic Staging for Non-Small Cell Lung Cancer Joe B. Putnam, Jr, MD Department of Thoracic Surgery, and the Department of Biomedical Informatics, Vanderbilt University, Nashville, Tennessee C linical staging techniques for non-small cell lung cancer (NSCLC) are commonly used to evaluate the patient for resection, and are based upon anatomic char- acteristics as a surrogate for biological aggressiveness and survival. The clinician defines the clinical stage—the best and final estimate of the extent of disease before the initi- ation of definitive therapy—for treatment recommenda- tions [1]. Despite significant improvements in preoperative clinical and invasive staging, occult N2 disease— metastasis to the mediastinal lymph nodes—continues to be identified. Intraoperatively, a systematic node dissection (SND), an integral component of every opera- tion for NSCLC, provides otherwise unachievable patho- logic staging information to guide subsequent treatment recommendations [2, 3]. For related article, see page 957 In this issue of The Annals, Obiols and colleagues [4] present their results of patients with occult N2 disease using a standardized clinical staging protocol and resec- tion with SND. Patients with histologically confirmed N2 disease by mediastinoscopy were excluded from subse- quent analysis. The investigators used the European So- ciety of Thoracic Surgeons (ESTS) guidelines for clinical staging [5]. Endobronchial ultrasonography (EBUS) was not available during this study period: invasive staging was with cervical mediastinoscopy. Approximately 75% of their patients underwent invasive staging before resec- tion, and 25% proceeded directly to resection without invasive staging. Obiols and colleagues [4] demonstrated a better than expected 3-year and 5-year survival of pa- tients with occult N2 disease after their preoperative staging algorithm and resection with SND: 3-year and 5- year survival was 79% and 40%, respectively, for patients with resected occult N2 disease, compared with interna- tional 5-year survival norms of 24% for pStage IIIA NSCLC [6]. Even with a structured preoperative staging strategy including mediastinoscopy, 5.5% of these pa- tients had unsuspected N2 disease. Patients, who pro- ceeded directly to surgery based on a negative clinical staging result, had a false negative rate of 7.5%, whereas patients who had previous invasive clinical staging of the mediastinum had a false negative rate of 5%. Should every patient have invasive staging of the mediastinum? Use of guidelines can direct patients with specific characteristics (suspicious mediastinal lymph nodes, clinical N1, central tumors, and so forth) toward invasive staging. In this study, there was only a small difference between the direct-to-surgery group and the surgically staged mediastinum group (7.5% and 5% false negative N2 rate, respectively). Selective mediastinoscopy seemed to be an effective strategy in this population, although only a minority of patients (25%) proceeded directly to resection without invasive surgical staging. Ancillary studies, including [18F]fluorodeoxyglucose positron emission tomography, performed relatively well in the noninvasively staged population (false negative rate 7.5%). With such a low number of false negative patients, should SND be abandoned or only limited to more advanced clinical stages? No—optimizing patho- logic staging even for early stage disease limits stage migration and most accurately guides subsequent treat- ment recommendations [7]. Most patients with occult N2 disease were subse- quently found to have only one lymph node station positive (typically, level 7 or 4R). Although many of these patients had clinical early stage disease, a SND was still performed and provided significant nodal tissue for analysis. It is interesting that for the 8 patients with 2 or more mediastinal lymph nodes, the 5-year survival was 0%. Granted, this is a small number of patients, but im- plies a greater negative impact on survival of patients with 2 or more mediastinal lymph nodes—even with complete resection. All patients in this study with path- ologically staged N2 disease had adjuvant chemotherapy as per current guidelines [8]. Endobronchial ultrasonography with transbronchial biopsy of the lymph nodes has more recently been applied as the initial invasive staging modality for patients with suspicious mediastinal lymph nodes; however, negative EBUS does not completely eliminate the need for cervical mediastinoscopy for patients with suspicious mediastinal lymph nodes nor does it limit the requirement for SND during the operation. In patients with a negative EBUS and suspicious lymph nodes by size, [18F]fluorodeoxyglucose avidity, or presence of clinical N1 disease, surgical staging of the mediastinum would still be appropriate [1]. Refinements in invasive staging such as EBUS will facili- tate the invasive clinical staging of the mediastinum, thereby improving treatment recommendations. In contrast to many practices of general thoracic sur- gery in the United States, these patients were treated with an open thoracotomy rather than with a minimally invasive or video-assisted thoracic surgery (VATS) approach. Analyses of aggregate data for the extent of lymph nodes dissected for VATS compared with open Address correspondence to Dr Putnam, Department of Thoracic Surgery, 609 Oxford House, 1313 21st Ave S, Nashville, TN 37232-4682; e-mail: [email protected]. Ó 2014 by The Society of Thoracic Surgeons Ann Thorac Surg 2014;97:744–6 0003-4975/$36.00 Published by Elsevier Inc http://dx.doi.org/10.1016/j.athoracsur.2013.12.040
Transcript of Disciplined Practice and Improving Clinical and Pathologic Staging for Non-Small Cell Lung Cancer
Disciplined Practice and Improving Clinical andPathologic Staging for Non-Small Cell Lung CancerJoe B. Putnam, Jr, MDDepartment of Thoracic Surgery, and the Department of Biomedical Informatics, Vanderbilt University, Nashville, Tennessee
linical staging techniques for non-small cell lung
Ccancer (NSCLC) are commonly used to evaluate thepatient for resection, and are based upon anatomic char-acteristics as a surrogate for biological aggressiveness andsurvival. The clinician defines the clinical stage—the bestand final estimate of the extent of disease before the initi-ation of definitive therapy—for treatment recommenda-tions [1]. Despite significant improvements in preoperativeclinical and invasive staging, occult N2 disease—metastasis to the mediastinal lymph nodes—continuesto be identified. Intraoperatively, a systematic nodedissection (SND), an integral component of every opera-tion for NSCLC, provides otherwise unachievable patho-logic staging information to guide subsequent treatmentrecommendations [2, 3].For related article, see page 957
In this issue of The Annals, Obiols and colleagues [4]present their results of patients with occult N2 diseaseusing a standardized clinical staging protocol and resec-tion with SND. Patients with histologically confirmed N2disease by mediastinoscopy were excluded from subse-quent analysis. The investigators used the European So-ciety of Thoracic Surgeons (ESTS) guidelines for clinicalstaging [5]. Endobronchial ultrasonography (EBUS) wasnot available during this study period: invasive stagingwas with cervical mediastinoscopy. Approximately 75% oftheir patients underwent invasive staging before resec-tion, and 25% proceeded directly to resection withoutinvasive staging. Obiols and colleagues [4] demonstrateda better than expected 3-year and 5-year survival of pa-tients with occult N2 disease after their preoperativestaging algorithm and resection with SND: 3-year and 5-year survival was 79% and 40%, respectively, for patientswith resected occult N2 disease, compared with interna-tional 5-year survival norms of 24% for pStage IIIANSCLC [6]. Even with a structured preoperative stagingstrategy including mediastinoscopy, 5.5% of these pa-tients had unsuspected N2 disease. Patients, who pro-ceeded directly to surgery based on a negative clinicalstaging result, had a false negative rate of 7.5%, whereaspatients who had previous invasive clinical staging of themediastinum had a false negative rate of 5%.
Should every patient have invasive staging of themediastinum? Use of guidelines can direct patients with
Address correspondence to Dr Putnam, Department of Thoracic Surgery,609 Oxford House, 1313 21st Ave S, Nashville, TN 37232-4682; e-mail:[email protected].
� 2014 by The Society of Thoracic SurgeonsPublished by Elsevier Inc
specific characteristics (suspicious mediastinal lymphnodes, clinical N1, central tumors, and so forth) towardinvasive staging. In this study, there was only a smalldifference between the direct-to-surgery group and thesurgically staged mediastinum group (7.5% and 5% falsenegative N2 rate, respectively). Selective mediastinoscopyseemed to be an effective strategy in this population,although only a minority of patients (25%) proceededdirectly to resection without invasive surgical staging.Ancillary studies, including [18F]fluorodeoxyglucosepositron emission tomography, performed relatively wellin the noninvasively staged population (false negativerate 7.5%). With such a low number of false negativepatients, should SND be abandoned or only limited tomore advanced clinical stages? No—optimizing patho-logic staging even for early stage disease limits stagemigration and most accurately guides subsequent treat-ment recommendations [7].Most patients with occult N2 disease were subse-
quently found to have only one lymph node stationpositive (typically, level 7 or 4R). Although many of thesepatients had clinical early stage disease, a SND was stillperformed and provided significant nodal tissue foranalysis. It is interesting that for the 8 patients with 2 ormore mediastinal lymph nodes, the 5-year survival was0%. Granted, this is a small number of patients, but im-plies a greater negative impact on survival of patientswith 2 or more mediastinal lymph nodes—even withcomplete resection. All patients in this study with path-ologically staged N2 disease had adjuvant chemotherapyas per current guidelines [8].Endobronchial ultrasonography with transbronchial
biopsy of the lymph nodes hasmore recently been appliedas the initial invasive staging modality for patients withsuspicious mediastinal lymph nodes; however, negativeEBUS does not completely eliminate the need for cervicalmediastinoscopy for patients with suspicious mediastinallymph nodes nor does it limit the requirement for SNDduring the operation. In patients with a negative EBUS andsuspicious lymph nodes by size, [18F]fluorodeoxyglucoseavidity, or presence of clinical N1 disease, surgical stagingof the mediastinum would still be appropriate [1].Refinements in invasive staging such as EBUS will facili-tate the invasive clinical staging of the mediastinum,thereby improving treatment recommendations.In contrast to many practices of general thoracic sur-
gery in the United States, these patients were treated withan open thoracotomy rather than with a minimallyinvasive or video-assisted thoracic surgery (VATS)approach. Analyses of aggregate data for the extent oflymph nodes dissected for VATS compared with open
Ann Thorac Surg 2014;97:744–6 � 0003-4975/$36.00http://dx.doi.org/10.1016/j.athoracsur.2013.12.040
745Ann Thorac Surg EDITORIAL PUTNAM2014;97:744–6 IMPROVING CLINICOPATHOLOGIC STAGING FOR NSCLC
techniques describe more variability in the number oflymph nodes resected with VATS (typically fewer) thanwith more traditional open techniques.
A recent single-institution study in The Annals evalu-ated the completeness of lymph node dissection or sam-pling for patients undergoing open lobectomy by openthoracotomy or VATS approach for clinical stage N0NSCLC. Lee and colleagues [9] found that the openapproach was superior to VATS in the mean number ofnodes dissected and the number of patients upstagedfrom N0 to pathologic N1 or N2. They identified no sur-vival differences between the two groups at 3 years, butexpressed concern regarding the adequacy of lymphnode dissection during VATS lobectomy [9].
A single-institution, propensity-matched study ofVATS and open procedure for patients undergoing lo-bectomy for NSCLC demonstrated more nodes (14.3versus 11.3; p ¼ 0.001) and more nodal stations (3.8 versus3.1; p<0.001) removed by open techniques compared withVATS. More than 90% of patients had clinical stage Idisease. Although VATS was not inferior to open pro-cedures with respect to overall and disease-free survival,the researchers recommended that open procedures maybe more appropriate for patients with more advancedclinical disease [10].
In another single-institution study for clinical stage 1patients, VATS approaches had fewer total number oflymph nodes dissected compared with open lobectomy,and fewer N2 nodes as well. Although no survival dif-ference was identified between the two groups, the in-vestigators recommended “more focused lymph nodesampling with VATS lobectomy” [11].
A broader evaluation of lymph node dissection wasexamined from The Society of Thoracic Surgeons generalthoracic surgery database of more than 11,500 clinicalstage I NSCLC patients undergoing operation [12]. Pa-tients undergoing an open approach were identified ashaving more occult nodal metastases than patients un-dergoing VATS. Patients with an open approach had astatistically significant N1 nodal upstaging, but no sig-nificant difference in upstaging from N0 to N2 metastasis,suggesting more variability in hilar and peribronchialdissection with a VATS approach.
Data from the National Comprehensive Cancer Net-work’s NSCLC database were analyzed for 388 patientswho underwent lobectomy (199 VATS and 189 open). Itwas found that open and VATS approaches were gener-ally similar in the percentage of patients who had at leastthree N2 stations examined, the number of N2 LN sta-tions, and the total number of N1þ N2 lymph nodes(although the median number was only four in bothgroups) [13].
Even with clinical early stage NSCLC, a SND facilitatespathologic staging, and may have a clinical benefit byreducing the variability of pathologic staging. Analysis ofsurgically treated NSCLC patients from the NationalCancer Database found that pStage I NSCLC patientswere best treated when a minimum of 10 lymph nodeswere resected at the time of any lung resection [14]. Pa-tients with fewer than 10 lymph nodes removed had
poorer survival (hazard ratio 1.21, 95% confidence inter-val: 1.18 to 1.25) compared with patients who had 10 ormore lymph nodes removed. Of interest is that 35% ofpatients had only four or fewer lymph nodes removed.Based on these data, the Commission on Cancer is eval-uating the following quality measure: “A total of atleast 10 lymph nodes are removed and pathologicallyexamined for resected NSCLC (pathologic stage IA, IB,IIA, and IIB).”The ACOSOG Z0030 trial (“Randomized trial of
mediastinal lymph node sampling versus complete lym-phadenectomy during pulmonary resection in the patientwith N0 or N1 [less than hilar] non-small cell carcinoma”)prospectively evaluated the therapeutic advantage oflymph node dissection (LND) compared with lymph nodesampling (LNS) in patients with early stage disease. Thestudy showed resection was safe [15], and demonstratedno difference in survival for patients with clinical earlystage NSCLC with LND, compared with patients whounderwent LNS [16]. The rate of occult N2 disease was 4%for patients with a negative LNS, and who were randomlyassigned to LND. The ACOSOG Z0030 study had astructured protocol for clinical staging, LNS, and LND.Despite no therapeutic survival advantage in the LNDgroup, the researchers emphasized the value of LND tooptimize pathologic staging in a patient with lung cancer.The choice of approach, either open or VATS, does notchange the need for LND even in patients with early stageNSCLC. Both open and VATS procedures were effectivein achieving the fundamentals of the operation, includingcomplete local control and lymph node dissection. Theseinvestigators observed that a median of 6 or more lymphnodes were obtained from at least three separate medi-astinal stations in 99% of patients [17]. They recom-mended that mediastinal lymphadenectomy shouldinclude stations 2R, 4R, 7, 8, and 9 for right-sided cancers;and stations 4L, 5, 6, 7, 8, and 9 for left-sided cancers [17].Future prospective clinical trials to evaluate anatomicstaging and its relationship to procedure outcome, andpatient survival, may be a lower priority than therapeutictrials, particularly with the explosion of molecular char-acterization of individual lung cancers. Larger populationstudies to compare the outcomes of patients with SND byopen and VATS techniques are needed.The quality and extent of lymph node dissection is at
the discretion of the surgeon. A systematic process forexamination and dissection of specific ipsilateral nodalstations is expected. The determination of number oflymph nodes versus number of lymph node fragmentswill be at the discretion of the pathologist. The volume (orweight) of lymph nodes resected has not been clearlyassociated with the veracity of pathologic staging. It isincumbent upon the surgeon to optimize intrathoracicstaging by a structured dissection of the hilar and medi-astinal nodal stations and removal of all accessible nodaltissues.In summary, patients with NSCLC should undergo a
structured preoperative clinical staging evaluationincluding surgical evaluation of the mediastinumwhere indicated to ensure optimal initial treatment
746 EDITORIAL PUTNAM Ann Thorac SurgIMPROVING CLINICOPATHOLOGIC STAGING FOR NSCLC 2014;97:744–6
recommendations, and complete resection of the tumorand a systematic node dissection to ensure the most ac-curate pathologic stage [18]. Guidelines from the NationalComprehensive Cancer Network [19], the ESTS [20], andthe American College of Chest Physicians [1] are prag-matic and outline a systematic approach to staging for thethoracic surgeon. Although the incidence of occult N2 islow (4% to 5.5%) in recent studies, complete resection inpatients with clinical N0 and subsequently pathologicoccult N2 is associated with better than expected survival.Is this related to a therapeutic effect from the SND orfrom improved pathologic staging? I cannot completelyanswer these questions from this manuscript—however,these patients were optimally selected, had the mostlimited N2 disease burden possible, and had a structuredapproach to both clinical staging and intraoperativelymph node dissection. Surgeons should not hesitate toproceed with resection to achieve optimal local control inthese patients. A systematic lymph node dissection is afundamental component of these procedures.
References
1. Silvestri GA, Gonzalez AV, Jantz MA, et al. Methods forstaging non-small cell lung cancer: diagnosis and manage-ment of lung cancer, 3rd ed: American College of ChestPhysicians evidence-based clinical practice guidelines. Chest2013;143(Suppl):e211–50.
2. Ludwig MS, Goodman M, Miller DL, Johnstone PA. Post-operative survival and the number of lymph nodes sampledduring resection of node-negative non-small cell lung can-cer. Chest 2005;128:1545–50.
3. Varlotto JM, Recht A, Nikolov M, Flickinger JC,DeCamp MM. Extent of lymphadenectomy and outcome forpatients with stage I nonsmall cell lung cancer. Cancer2009;115:851–8.
4. Obiols C, Call S, Rami-Porta R, et al. Survival of patients withunsuspected pN2 non-small cell lung cancer after an accu-rate preoperative mediastinal staging. Ann Thorac Surg2014;97:957–64.
5. Gunluoglu MZ, Melek H, Medetoglu B, Demir A, Kara HV,Dincer SI. The validity of preoperative lymph node stagingguidelines of European Society of Thoracic Surgeons in non-small-cell lung cancer patients. Eur J Cardiothorac Surg2011;40:287–90.
6. Goldstraw P, Crowley J, Chansky K, et al. The IASLC lungcancer staging project: proposals for the revision of the TNMstage groupings in the forthcoming (seventh) edition of theTNM classification of malignant tumours. J Thorac Oncol2007;2:706–14.
7. Al-Sarraf N, Aziz R, Gately K, et al. Pattern and predictors ofoccult mediastinal lymph node involvement in non-smallcell lung cancer patients with negative mediastinal uptakeon positron emission tomography. Eur J Cardiothorac Surg2008;33:104–9.
8. Ramnath N, Dilling TJ, Harris LJ, et al. Treatment of stage IIInon-small cell lung cancer: diagnosis andmanagement of lungcancer, 3rded:AmericanCollege ofChest Physicians evidence-based clinical practice guidelines. Chest 2013;143(Suppl):e314–40.
9. Merritt RE, Hoang CD, Shrager JB. Lymph node evaluationachieved by open lobectomy compared with thoracoscopiclobectomy for N0 lung cancer. Ann Thorac Surg 2013;96:1171–7.
10. Lee PC, Nasar A, Port JL, et al. Long-term survival after lo-bectomy for non-small cell lung cancer by video-assistedthoracic surgery versus thoracotomy. Ann Thorac Surg2013;96:951–60.
11. Denlinger CE, Fernandez F, Meyers BF, et al. Lymph nodeevaluation in video-assisted thoracoscopic lobectomy ver-sus lobectomy by thoracotomy. Ann Thorac Surg 2010;89:1730–5.
12. Boffa DJ, Kosinski AS, Paul S, Mitchell JD, Onaitis M. Lymphnode evaluation by open or video-assisted approaches in 11,500 anatomic lung cancer resections. Ann Thorac Surg2012;94:347–53.
13. D’Amico TA, Niland J, Mamet R, Zornosa C, Dexter EU,Onaitis MW. Efficacy of mediastinal lymph node dissectionduring lobectomy for lung cancer by thoracoscopy and tho-racotomy. Ann Thorac Surg 2011;92:226–31.
14. Pezzi CM, Gay EG, Kulkarni N, Putnam JB. Limited thoraciclymphadenectomy worsens survival in 55,122 patients withresected stage I non-small cell lung cancer. Ann Surg Oncol2013;20(Suppl 1):105.
15. Allen MS, Darling GE, Pechet TT, et al. Morbidity andmortality of major pulmonary resections in patients withearly-stage lung cancer: initial results of the randomized,prospective ACOSOG Z0030 trial. Ann Thorac Surg 2006;81:1013–9.
16. Darling GE, Allen MS, Decker PA, et al. Randomized trial ofmediastinal lymph node sampling versus complete lym-phadenectomy during pulmonary resection in the patientwith N0 or N1 (less than hilar) non-small cell carcinoma:results of the American College of Surgery Oncology GroupZ0030 trial. J Thorac Cardiovasc Surg 2011;141:662–70.
17. Darling GE, Allen MS, Decker PA, et al. Number of lymphnodes harvested from a mediastinal lymphadenectomy: re-sults of the randomized, prospective American College ofSurgeons Oncology Group Z0030 trial. Chest 2011;139:1124–9.
18. Howington JA, Blum MG, Chang AC, Balekian AA,Murthy SC. Treatment of stage I and II non-small cell lungcancer: diagnosis and management of lung cancer, 3rd ed:American College of Chest Physicians evidence-based clin-ical practice guidelines. Chest 2013;143(Suppl):e278–313.
19. Ettinger DS, Wood DE, eds. National Comprehensive CancerNetwork.NCCNclinical practice guidelines in oncology.Non-small cell lung cancer version 2.2014.Available at: http://www.nccn.org/professionals/physician_gls/pdf/nscl.pdf. AccessedDecember 26, 2013.
20. De Leyn P, Dooms C, Kuzdzal J, et al. Revised ESTS guide-lines for preoperative mediastinal lymph node staging fornon-small cell lung cancer. Available at: http://www.ests.org/guidelines_and_evidence/ests_guidelines.aspx. AccessedJuly 16, 2013.
