Lung Cancer

View TSDA Curriculum Online for this topic

4. Clinical Manifestation of Lung Cancer

Cough (most common 75%), hemoptysis (33%), pain (50% poor prognostic signs), anorexia and weight loss (poor prognostic sign), shortness of breath, pleural effusion, hoarseness (1-8%)
Cushing's syndrome (most common), inappropriate ADH secretion, eosinophilia (tumor necrosis) and neuromyopathies (15%)

  Full TNM Staging Classification

TX   Primary tumor cannot be assessed, or tumor proven by the presence of 
         malignant cells in sputum or bronchial washings but not visualized by 
         imaging or bronchoscopy

 T0    No evidence of primary tumor

 Tis   Carcinoma in situ

 T1   Tumor 3 cm or less in greatest dimension surrounded by lung or visceral 
         pleura, without bronchoscopic evidence of invasion more proximal than
         the lobar bronchus (not in the main bronchus)

 T2    Tumor with any of the following features of size or extent:

More than 3 cm in greatest dimension
Involves main bronchus, 2 cm or more distal to the carina
Invades the visceral pleura
Associated with atelectasis or obstructive pneumonitis that extends to the hilar region but does not involve the entire lung

 T3    Tumor of any size that directly invades any of the following:  chest wall 
         (including superior sulcus tumors), diaphragm, mediastinal pleura, 
         parietal pericardium; or tumor in the main bronchus less than 2 cm distal
         to the carina, but without involvement of the carina; or associated 
         atelectasis or obstructive pneumonitis of the entire lung

 T4   Tumor of any size that invades any of the following:  mediastinum, heart,
        great vessels, trachea, esophagus, vertebral body, carina; or separate
        tumor nodules in the same lobe; or tumor with a malignant pleural
        effusion.

 NX Regional lymph nodes cannot be assessed

 N0 No regional lymph node metastasis

 N1 Metastasis to ipsilateral peribronchial and/or ipsilateral hilar lymph nodes
       and intrapulmonary  nodes including involvement by direct extension of 
       the primary tumor

 N2 Metastasis to ipsilateral mediastinal and/or subcarinal lymph node(s)

 N3 Metastasis to contralateral mediastinal, contralateral hilar, ipsilateral or 
       contralateral scalene, or supraclavicular lymph node(s)

 MX Distant metastasis cannot be assessed

 M0 No distant metastasis

 M1 Distant metastasis (includes synchronous separate nodule(s) in a 
 different lobe)

Important for prognosis and therapy.

10. Complete Pulmonary Resection

Surgeon is morally certain he or she has encompassed all tumor disease
Proximal margins of resected specimen are microscopically free of tumor
Within each major lymphatic drainage region, the most distal node is microscopically free of tumor
Capsules of resected nodes are intact.

Only 50% of patients with lung cancer are surgical candidates
50% of surgical patients have mediastinal N2 disease
Mediastinal lymph node dissection is:

Only definitive way of staging lung cancer
Identifies patients with skip metastasis (33% incidence especially in adenocarcinoma)
Identifies intranodal V/s perinodal metastasis
Identifies multilevel disease (poor prognosis)
Is required by many neoadjuvant and adjuvant protocols
Is a part of a complete resection

Current imaging techniques determine size not histology
Malignant mediastinal nodes are not larger than benign lymph nodes 
(58% > 15 mm benign)
Small mediastinal nodes (< 10 mm) are not infrequently malignant (15%)
Benign adenopathy is more common in patients with acute pulmonary inflammation
Pathologic confirmation rates higher than radiologic estimation

CT identification of enlarged hilar/mediastinal nodes is not diagnostic of advanced stage disease. CT, therefore adds an extra cost without contributing to the management plan

 

Stage IA (511)  Stage IB (549)  Stage IIA (76)  Stage IIB (375)  Stage IIIA (399)

- 67% - 57% - 55% - 39% - 23%

Factors influencing survival in N2 disease

Multiple levels of involvement
Nodal vs extranodal disease
Superior vs inferior mediastinum
Bulky clinical vs discrete CT nodes

Recurrences - 80% within 2 years
Second primary 3-4% per year, especially in high risk patients

Distant metastases (absolute)
Malignant pleural effusion (absolute)
Superior vena caval syndrome
Horners syndrome
Vocal cord paralysis
Phrenic nerve paralysis

Neoadjuvant therapy has been successful in anal, bladder and esophageal cancers. Head and neck cancers do not respond to neoadjuvant therapy

Rationale for neoadjuvant therapy

Surgical resection disrupts blood supply and adjuvant therapy may not be deliverable
Preoperative therapy may minimize seeding
Preoperative therapy may accomplish downstaging
Tumor growth is inversely related to size. Micrometasteses grow faster 
Chemotherapy related killing follows first order of kinetics
Goldie Coldman hypothesis - with each cell division cells become resistant due to continued mutation

17. Current Neoadjuvant Therapy Trials
 

18. Conclusions from randomized trials

Well tolerated with high response and resectability rates (70%)
Trend towards increased disease free survival and even overall survival

19. Chest Wall Invasion of Lung Cancer

Incidence of about 5%
Does not imply a hopeless prognosis
Mortality and morbidity for surgical resection are acceptable
Extensive resections possible

VI. Epidemiology 

A. World 
1. 900,000 people/yr 
2. Most common cause of death for men 
3. Incidence will rise through the 1990’s 

B. USA 
1. More cancer deaths for men and women than any other 
2. 10th in world 

C. Men 
1. Increased from 5/100,000 to 57/100,000 (1930’s to 1990) 
2. Death rates declining for younger men (<55 yo), increasing in older 

D. Women 
1. 1965-1990 = 8-fold increase in lung cancer incidence in women 
2. Leading cause of cancer death age 55-74 
3. More likely to be non-smokers 

E. Young adults 
1. < 5% are < 40 yo 
2. More aggressive - adeno and small-cell more common (than in older pts) 
3. M:F= 1.7:1 vs 3.47:1 for older pts 

VII. Etiology 

A. Tobacco 
1. 90% of cancers are in smokers 
2. Latent period  20-25 years 
3. Dose-related (heavy smoker 25x risk of non-smoker) 
4. “Most serious drug addiction in the world” 

B. Components of cigarette smoke 
1. Tumor initiators 
a) ³ 2 dozenpolycyclic aromatic hydrocarbons 
2. Tumor promoters (or cocarcinigens) 
a) Fatty acids, phenols, N-methylated indoles, insecticides 
b) Low-dose nicotine 
c) N-nitrosamines 
3. Complete carcinogens 
a) Nickel, arsenic 
b) Radioactive plutonium (80 pk-yr = 1300 rem from polonium) - enough to cause Ca 

C. 2nd hand smoke 
1. Side-stream smoke more carcinogenic per weight 
2. Respiratory symptoms and illnesses in kids 
3. Risk enhanced 25% in non-smoking “passive smokers” 

D. Industrial exposure 
1. Arsenic, chromates, nickel, asbestos, silica, iron, coal 
2. Organic chemicals: benzopyrene, vinyl chloride, chloromethyl ether 
3. Radioactive emissions 
4. Newspaper workers, miners, halothane workers 
5. Asbestos 
a) Synergy w/tobacco (50-90x risk of Ca c/w control 
b) 1/5 deaths from bronchogenic Ca, 1/10 from pleural/peritoneal mesothelioma, 1/10 from GI Ca 

E. Atmospheric pollution 
1. Radon 
2. Chernobyl (Hungary is #1) 

F. Genetic factors 
1. Specific “lung cancer genes” not yet identified 
2. ras-  and myc- oncogenes associated w/ growth regulation 

VIII. Histogenesis and Pathogenesis of lung Ca 

A. Background 
1. Stem cell theory - cells may differentiate and lose differentiation 
Small Cell 
Adenocarcinoma
Squamous Cell Carcinoma 

2. ras, myc or neu oncogene mutations may be associated with a particular tumor type 
3. Pulmonary epithelial cells - limited repertoire of response 
a) Basal cell hyperplasia with metaplasia & differentiation toward goblet or squamous cell 
b) Proliferation of Kulchitsky cells 
c) Proliferation of type II pneumocytes (stem cell, produces surfactant) 

B. Stages of Carcinogenesis 
1. Initiation - irreversible, rapid, 2° to genotoxin 
2. Promotion - reversible, clonal expansion of initiated cells 
3. Progression, invasion, metastasis - less well understood 
4. nm23 gene - high expression associated w/ low metastatic potential (murine model) 

C. Squamous carcinogenesis - (better defined stages than adenocarcinogenesis) 
1. Hyperplasia 
a) ­ growth factors à ­ # of basal (reserve) cells in bronchial epithelium 
b) Cells are benign, respond to normal control mechanisms 
2. Metaplasia 
a) Reversible 
b) Ciliated bronchial epithelial cells àgoblet or squamous cells 
3. Dysplasia 
a) Reversible 
b) Thickened epithelium 
c) Cellular orientation and maturation disordered, but present 
d) Cytologically abnormal with ­ Nuc/Cyt ratio 
e) Quantitative DNA analysis may risk stratify degrees of dysplasia 
4. Carcinoma in situ 
a) Thickened, severely atypical epithelium 
b) Cytologic features similar to, but more severe than dysplasia 
c) Similar to carcinoma, but no invasion of basement membrane 
d) Several foci frequently seen outside of tumor in resected specimens 

IX. Staging - see American Joint Committee for Cancer Staging and End Results Reporting (AJC) and Union Internationale Contre Cancer (UICC) nomenclature, next page 

X. Classification of lung tumors - see text for histology 

A. Usual tumors constitute 95% 

B. 45% have mixed histologic pattern by light microscopy 

XI. Squamous cell carcinoma (SCC) 

A. 90% in segmental or larger bronchi 

B. Grow endobronchially and invade peribronchial soft tissue 

C. Central tumors can be dx’d by sputum cytology - most common neoplasm detected in screening program 

D. Typically metastasize within thorax 
1. Pleura involved in 1/3 - cells in pleural effusion uncommon 
2. Only 20-25% have extra-thoracic mets 

E. Peripheral SCC 
1. Keratin pools 
2. Cavitate 

F. Concomitant upper and lower airway SCC 
1. Poor prognosis 
2. Stage I, II, asymptomatic, may be cured 
3. Segmentectomy due to poor pulm. fxn and likelihood of further Ca 

XII. Adenocarcinoma 

A. Incidence 
1. Increasing 
2. Most common in Japan and in females in USA 

B. 30% arise in surface epithelium and submucosal glands of bronchi smaller than in SCC 

C. Non-ciliated bronchiolar epithelial cell (Clara cell) may be common cell of origin 

D. Appearance 
1. Hard, gray or white mass in periphery 
2. Necrosis can occur, cavitation is rare 
3. Desmoplastic response 

E. Grow more rapidly than SCC 

F. Metastasize early - > 80% present w/mets (adrenal, liver, bone, brain) 

G. Presentation 
1. Asymptomatic mass in periphery 
2. FNA most common method of dx 

H. Subtypes - most show more than one subtype 
1. Acinar - most common 
2. Papillary 
3. Mucin-producing 

XIII. Bronchoalveolar (Bronchiolar) Carcinoma (BAC) 

A. Incidence increasing, 9.3-20.3% from 1978-1989 

B. Clinical presentation 
1. Arises in periphery 
2. Tends to be multicentric 
3. Radiographically, may be an infiltrate rather than a mass, may not change for years 
4. Relationship to smoking debated 

C. Segmentectomy, when possible, as end stage is frequently respiratory failure 

D. Histology 
1. Subtype of adeno 
2. Clara cells and type II pneumocytes are cells of origin 
3. Grow along alveolar septa 
4. 3 morphologic types 
a) Well-differentiated tall columnar 
b) Tall columnar w/”hobnail” caps 
c) Cuboidal 

XIV. Large Cell undifferentiated 

A. 50% arise in subsegmental or larger bronchi (between SCC and adeno) 

B. Aggressive - often stage III or IV,  poor prognosis 

XV. Combination carcinomas 

A. 45-50% of all Ca 

B. Assume it will behave like most malignant component 

XVI. Scar carcinoma 

A. 1/3 adeno, 1/5 SCC have sufficient scar around tumor to raise ? 

B. 50% assoc. w/healed infarcts or arrested granulomatous inflammation, 25% w/non-specific inflammation 

C. Most often in upper lobes, creating diagnostic challenge 
1. Is excision of a long-standing scar warranted if there is a slight radiographic D? 
2. FNA of a “new” mass yields scar, is that representative? 

XVII. Neuroendocrine (Kulchitsky Cell) Carcinomas 

A. Background 
1. APUD (Amine precursor uptake derivative) origin 
2. First evidence that DNA ploidy correlates with aggressiveness 
3. Common cell of origin for typical and atypical carcinoid tumors 
4. 10% thought to have carcinoid have had Small-cell 
5. Small-cell carcinomas included in this group (a genetic variant of carcinoid) 

B. Carcinoid: Typical (KCC I) and Atypical (KCC II) 
1. 3-5% of all lung Ca 
2. 90% are central, arising in lobar or segmental bronchi, and grow in a polypoid manner 
3. Covered by an intact mucosa, most are 2-4cm at time of dx 
4. 5% metastasize - usually to regional nodes 
5. Typical carcinoid (KCC I) 
a) May manifest w/hemoptysis and nl CXR 
b) Bx (bronchoscopy), then complete resection (sleeve resection useful)) 
c) Laser ablation not recommended, as tumor is deep to mucosa 
d) Carcinoid syndrome <3% 
e) 94-100% 5-yr survival 
6. Atypical carcinoid - 10%  (KCC II) 
a) 46% are stage II or III 
b) 20% have M1 dz 
c) Mean survival 25 months 
d) Dx may be confused w/SCLC 
7. Spindle cell carcinoid -resembles smooth muscle cell

C. Small Cell undifferentiated Carcinoma 
1. Arise from basal cells of bronchial epithelium 
2. Neuroendocrine differentiation 
3. Metastasize early and widely 
a) Hilar adenopathy 
b) Brain 
4. With treatment, may “mature” to SCC 
5. Need adequate bx specimen to establish dx (vs atypical carcinoid, lymphoproliferative dz) 
6. Stage I may be tx’d w/resection + chemo tx (?Stage II) 
7. Histology 
a) Oat cell 
b) Intermediate cell 

D. Large Cell Neuroendocrine Carcinoma - behaves between KCC II and SCLC 

XVIII. Other neoplasms of respiratory tract 

A. Adenoid cystic carcinoma (cylindroma) 
1. Mainstem or major bronchus 
2. Complete excision can result in cure, but local late recurrence possible 
3. Histology similar to salivary gland 

B. Mucoepidermoid carcinoma 
1. Low-grade, slow-growing 
2. Glandular and/or squamous differentiation 

C. Carcinosarcoma 

D. Pulmonary blastoma, fibrosarcoma, hemangiopericytoma 

E. Pulmonary lymphoma - resection is tx of choice 

XIX. Paraneoplastic syndromes: see chart 

XX. Diagnostic pitfalls 

A. FNA - sampling error 

B. Bronchoscopy 
1. Obtain several specimens 
2. Await permanent sections due to small specimens 

C. Fungal infection notorious for cellular atypia (mimic cancer) 
 

Surgical Treatment of Lung Carcinoma

VI. Extent of resection 

A. 1933 - Graham and Singer - First pneumonectomy for carcinoma 

B. 1950 - Churchill - Lobectomy is effective for Ca and safer than pneumonectomy 

C. Less than lobectomy 
1. Jensik - Peripheral Stage I (n=168) 
a) Wedge or segmentectomy 
b) 53% 5-year 
c) 45 pts died of disease, 16/45 - local recurrence 
2. In retrospective series, loco-regional recurrence: 4.4-22.7% vs 4.9-11.5% 
3. Ginsberg & Rubinstein 
a) Randomized: lesser resection vs lobectomy 
b) Loco-regional recurrence: 17.2% vs 6.4% 
c) 5-yr survival: 68% vs 50% 
4. Recommendation (Glenn’s): May be useful for high-risk, elderly pt with limited reserve 

D. Incomplete resection 
1. 1-yr = 26%, 3-yr = 8.5%, 5-yr = 4% 
2. All 5-yr survivors had SCC with + bronchial margin 

E. Mediastinal lymph node dissection - necessary for pathologic + surgical staging 
1. Superior mediastinal/ paratracheal - Right 
a) Anterior mediastinum (ant to SVC) nodes resected if palpable 
2. A-P window - Left 
a) Supraaortic and superior mediastinal palpable nodes excised 
3. Subcarinal & lower mediastinal - Bilateral 

F. VATS 
1. Controversial 
2. Criteria for VATS resection 
a) Stage I 
b) £ 2cm 
c) Lower lobes preferable 
d) Incomplete fissure makes resection difficult 

VII. Occult non-small cell lung carcinoma (NSCLC) = TX N0 M0 

A. + sputum cytology on screening or hemoptysis without radiographic manifestations 

B. 1/3 with + sputum  represent head and neck primary 

C. FOB if head and neck exam is normal 
1. If appearance is normal,  do segmental brushings and cytology 

D. Tx is lobectomy or pneumonectomy 
1. Median survival 9 years 
2. 45% develop new carcinomas usually airway 

E. 1/3 have mediastinal and/or lymphatic spread 

F. Screening 
1. Low yield (53/10,040 smokers) 
2. In-vivo fluorescence - Ý sensitivity of brocnhoscopy in screening and in TX cases 

VIII. Stage I NSCLC - T1 N0 M0, T2 N0 M0 

A. Older staging included T1 N1 M0 as stage I 

B. ~ 20% of patients 

C. T1 = £ 3cm, surrounded by lung or visceral pleura 

D. T2 = >3cm or any size w/ visceral pleural invasion or atelectasis extending to hilum and >2cm distal to carina 

E. Staging 
1. H&P 
2. Lab: SGOT, Af, LDH 
3. CT chest + upper abd to include liver and adrenals 
4. Without symptoms, bone and brain scan unnecessary (controversial) 

F. Lobectomy, bilobectomy or pneumonectomy + mediastinal LND 
1. OR mortality = 0-2.3% 
2. 5-yr survival 63-85% (75%) 
3. Prognostic factors 
a) NOT- age, sex, pleural involvement, grade, histologic type 
b) DNA ploidy & grade - Ichinose 
4. 39% develop recurrence or new lung 1° (5-10 yr) 
a) 56% distant mets 
b) 20% new lung 1° 
5. F/U = CXR + PE q 3months x 1st yr, q4mo 2nd yr, then q 6mo 

G. No adjuvant tx recommended (?chemoprevention) 

IX.  Stage II NSCLC - T1 N1 M0, T2 N1 M0 

A. 10% of pts 

B. T1,2 as above (III B,C) and peribronchial or ipsilateral nodes 

C. Lobectomy, bilobectomy or pneumonectomy + mediastinal LND 
1. 39-49% 5- yr survival 
2. 40-54% (T1N1) / 38-40% (T2N1) 
3. Prognostic factors 
a) ­ tumor size and ­ # of nodes 
b) Not -age sex, pleural involvement 
c) ? histology (SCC better) 
4. Recurrence - 55% 
a) 21% loco-regional 
b) 79% distant (47% brain mets) 
c) SCC à loco-regional / Adenoà distant 

D. Adjuvant Tx 
1. Chemo-tx - role unclear, not recommended 
2. Rad tx à ß loco-regional recurrence, no D in survival 
3. Post-op immuno-tx not beneficial 

X. Stage IIIA NSCLC 

A. New international staging 
1. Adds stage IV for distant mets (M1) 
2. Divides Stage III 
a) IIIA  (T3 N0-1 M0, T1-3 N2 M0) 
(1) Limited extrapulmonary extension, or ipsilateral or subcarinal nodes 
(2) Pts may be offered surgical resection 
b) IIIB (Any T N3 M0, T4 any N M0) 
(1) Pts should be considered for surgery only in a special protocol setting 

B. T3 - Chest wall invasion (Excluding superior sulcus tumors) 
1. 5% of pts 
2. Surgical Tx 
a) Pulmonary resection 
(1) Extrapleural if parietal pleural involvement noted at thoracotomy and tumor-free plane exists 
(2) En-bloc if plane cannot be achieved 
b) Soft tissue resection 
c) Mediastinal LN dissection 
d) Chest wall reconstruction 
(1) Chapter 34 
(2) Methylmethacrylate/Marlex sandwich 
3. Operative mortality 4-12% 
4. 5-year survival 26-40% 
5. Prognosticators (poor) 
a) Incomplete resection - median 9 mo survival, 0 3-yr survival 
b) LN involvement (N1-2) - 21% vs 56% 
c) Chest wall vs parietal pleura alone 16% vs 48% 
6. Radiation 
a) No randomized, controlled data 
b) Patterson (retrospective) - 56% vs 30% 5-yr 

C.  T3 - <2cm from carina 
1. Bronchoscopy 
a) Proximity to carina 
b) Submucosal spread 
2. Sleeve lobectomy 
a) 30-64% 5-yr survival 
b) 0-8% operative mortality (pneumonectomy - 6% operative mortality) 
c) By stage: 
(1) I - 38% 5-yr 
(2) II- 20% 
(3) III- 15% 
3. Sleeve pneumonectomy 
a) Bulky central tumor in proximity to or involving carina or tracheobronchial angle 
b) 4-31% operative mortality 
c) Anastomotic dehiscence à 100% mortality 
d) 16-23% 5-yr survival 

D. N2 dz 
1. 45% of pts 
2. Some feel N2 is not resectable (with or without RT/Chemo) 
3. Pearson’s mediastinoscopic contraindications to resectability 
a) Contralateral nodal dz 
b) Extranodal extension 
c) High paratracheal nodal dz 
4. Survival = 20-30% 5-yr survival without pre-op mediastinoscopy, with post-op RT 
5. 706 pts w/N2 dz (see table below) 
a) T1 - 46% 5-yr survival 
b) T2 - 27% 
c) T3 - 14% 
d) Level 1, 2  nodes (upper paratracheal) à 20% vs 32 % 
e) Level 7 (subcarinal) à 22% vs 33% 
f) A-P window à 35% 
6. Adjuvant tx 
a) LCSG - Post-op RT vs no RT 
(1) No survival benefit for stage II, III epidermoid 
(2) ­ loco-regional control 
b) LCSG - Adeno & Large-cell post-op immuno (BCG +levamisole) vs Chemo (Cytoxan, adriamycin, cisplatin) 
(1) Survival similar 
(2) Dz-free survival better w/chemo-tx 
c) Roth, Rosell - pre-op Chemo vs surgery alone in IIIA 
(1) 3-yr = 30, 56% vs 0, 15% 
d) Recommendation: Pre-op chemo-tx should be offered as part of a trial 
 Clinical N2 (CXR, Bronchoscopy, +/- mediastinoscopy) Clinical N1,N0 (CXR, Bronch) 
Complete Resection 18% 53% 
3-yr survival 9% 47% 
5-yr survival 9% 37% 

VI. IIIB 

A. T4 - Pleural effusion 
1. Cytology is usually positive à median 6-9 month survival 
a) Palliative tx - Chest tube, then chemical pleuradesis (80%) 
b) Subtotal pleurectomy for failures or incomplete re-expansion 
2. Cytology negative, non-bloody, not an exudate - exclude this pleural effusion for staging purposes 
3. Evaluation 
a) Thoracentesis 
b) Thoracoscopy w/bx of pleural lesions 

B. T4 - Mediastinum (n=225) 
1. Complete resection 
2. Incomplete resection + brachytherapy 
3. Brachytherapy alone 
4. Incomplete or no resection w/o brachytherapy 
5. Overall, 22%, 13%, 7% at 2,3,5 yrs 
6. Consider adjuvant tx 

VII. Stage IV M1 NSCLC 

A. Brain mets 
1. 27-48% in autopsy series 
2. 47% of pts w/M1 dz 
3. Survival from onset of symptoms = <1 month to 6 months 

B. Resection of solitary brain metastasis 
1. Operative mortality 2-44% 
2. 2.6-12 mo survival 
3. Retrospective data 
a) Surg + RT vs RT alone à 16 vs 4 mo survival 
b) 13% 5-yr survival, median 14 mo survival 
4. Prospective (Patchell) 
a) Surg + whole brain RT vs whole brain RT alone 
b) 9.2 vs 3.4 mo median survival 
5. Recommendation 
a) Synchronous, solitary brain met, no other mets àcraniotomy, then thoracotomy 
b) Brain met discovered after lung resection àcraniotomy 
c) RT after resection of brain met controversial 

VIII. Small Cell Carcinoma 

A. Most patients have abnormal mediastinum, \ bx for dx, then RT + Chemo tx 

B. Small-cell discovered at thoracotomy 
1. Stage I àresect 
2. 5-yr 60/28% survival for T1/T2 
3. Chemo tx recommended 
4. Stage II, III àRT + Chemo (no resection) 

IX. Closing comments 

A. 2% operative mortality 

B. High-risk (consider lesser resection) 
1. > 70 yr old for major resection 
2. Cardiovascular dz 
3. Poor lung fxn