Radiation Oncology: A Question-Based Review

Early-Stage (I-II) Non-Small Cell Lung Cancer

John P. Christodouleas and Joe Y. Chang

image Background

How many lobes are there in the lung? How many segments are there per lobe?

There are 5 lobes in the lung–3 on the right and 2 on the left: RUL, RML, RLL, LUL, and LLL. Lingula is the anatomic equivalent of LML, and is part of the LUL. There are 5 segments per lobe, except for the RUL and RML, which are divided in 3 and 2 segments, respectively, supplied by tertiary bronchi.

Name the 9 N2 nodal stations.

N2 nodal stations:

1.     Station 1: highest mediastinal

2.     Station 2: upper paratracheal

3.     Station 3: prevascular (3A) and retrotracheal (3P)

4.     Station 4: lower paratracheal

5.     Station 5: subaortic (AP window)

6.     Station 6: para-aortic

7.     Station 7: subcarinal

8.     Station 8: paraesophageal

9.     Station 9: pulmonary ligament

Where are the intrapulmonary and hilar nodes located?

Intrapulmonary nodes are nodes along the secondary bronchi, whereas hilar nodes are those along the main stem bronchi. These are all considered N1 nodes.

Name the 5 N1 nodal stations.

N1 nodal stations:

1.     Station 10: hilar

2.     Station 11: interlobar

3.     Station 12: lobar

4.     Station 13: segmental

5.     Station 14: subsegmental

(Note: N1 nodes are all double digits.)

What are the estimated annual # of new lung cancer cases diagnosed in the U.S. and the # of deaths from lung cancers?

In 2009, there were ~219,000 newly diagnosed cases of lung cancers in the U.S., accounting for > 160,000 deaths. This accounts for more deaths than all colorectal, breast, and prostate cancers combined.

Overall, what is the 5-yr survival rate for lung cancer pts?

The overall 5-yr survival rate for non–small cell lung cancer (NSCLC) is 15%.

What are the 3 histologic subtypes of NSCLC in decreasing order of frequency?

Histologic subtypes of NSCLC: adenocarcinoma (50%) > squamous cell carcinoma (35%) > large cell (15%)

In addition to tobacco smoke, what are 3 other environmental exposure risk factors for developing lung cancers?

Environmental exposure risk factors for lung cancer:

1.     Radon

2.     Asbestos (Note: Smoking and asbestos exposures are synergistic in early reports, but more recent studies suggest less than a multiplicative effect.)

3.     Occupational exposure (arsenic, bis-chloromethyl ether, hexavalent chromium, mustard gas, nickel, polycyclic aromatic hydrocarbon)

What is the estimated RR for lung cancer in heavy smokers vs. nonsmokers?

Heavy smokers have a 20-fold excess of lung cancer (American Cancer Society [ACS] cohort study).

What is the risk of lung cancer in former smokers compared to current smokers?

The risk of developing lung cancer in former smokers is around half (9 times vs. 20 times) that of current smokers (ACS cohort study).

What is the risk of lung cancer from passive smoke exposure?

There is an RR of 1.2–1.3 for developing lung cancer from passive smoke exposure.

Approximately what % of smokers develop lung cancer?

<20% of smokers actually develop lung cancer (in the Carotene and Retinol Efficacy Trial, 10-yr cancer risk was 1%–15%).

What histology subtype of NSCLC is least associated with smoking?

Adenocarcinoma is the histologic subtype that is least associated with smoking.

Name 3 histologic variants of adenocarcinoma of the lung.

Bronchoalveolar, acinar, and papillary

Name 2 variants of large cell cancer of the lung.

Giant cell and clear cell

What is the race and gender predilection for NSCLC?

Blacks have the highest incidence of lung cancer. Males also are historically at greater risk, but as females continue to start smoking, the incidence in females is rising.

What is the most common stage at initial presentation?

The most common stage of presentation for lung cancer is metastatic Dz (around one third of pts).

What are the most common sites of DMs for lung cancer?

Bone, adrenals, and brain

What are the para-neoplastic syndromes associated with lung cancers?

Hypercalcemia of malignancy due to PTHrP, syndrome of inappropriate secretion of antidiuretic hormone → ↓Na, Cushing, Lambert-Eaton syndrome, and other neurologic disorders

What is the cause of Lambert-Eaton syndrome? Clinically, how can Lambert-Eaton be distinguished from myasthenia gravis?

Lambert-Eaton syndrome is caused by circulating autoantibodies against presynaptic P/Q calcium channel. Lambert-Eaton strength improves with serial effort, but not myasthenia gravis.

Which histologic subtypes of lung cancer are associated with peripheral and central locations?

1.     Peripheral: adenocarcinoma

2.     Central: SCC

With which histologic subtypes of lung cancer is thyroid transcription factor-1 (TTF-1) staining associated?

Adenocarcinoma, nonmucinous bronchioalveolar carcinoma, and neuroendocrine tumors (i.e., small cell lung cancer, carcinoid). TFF-1 is rare in SCC. A thyroid cancer primary must be excluded.

In NSCLC, what is the role of CT screening for high-risk pts?

This is controversial as of 2010. Lead time bias could be the reason why survival is better. IELCAP (Henschke CI et al., NEJM 2006) reported that out of 27,456 pts screened, 74 pts were found to have cancer (0.3% detection) and 86% were stage I. 10-yr survival was 93% in stage I pts who underwent resection at Dx. 10-yr OS was 82% for all pts diagnosed by CT.

What is the single most clinically significant acquired genetic abnormality in NSCLC?

EGFR mutation in exon 19 (in-frame deletion of 4 aa, LREA) and exon 21 (L858R point mutation); results in a constitutive active receptor.

Among pts with NSCLC, in what particular groups are the EGFR mutations common, and for what do these mutations predict?

In the overall lung cancer population, EGFR mutations are seen in only ~10%, but this occurs at high rates (30%–70%) in nonsmokers, adenocarcinomas, and Asians. These mutations predict for a high response rate to TKIs (gefitnib, erlotnib) of ~80%.

What point mutation in the EGFR gene is associated with TKI resistance?

T790M is the point mutation in the EGFR gene associated with TKI resistance.

What other genetic alteration predicts well for response to TKI?

EGFR amplification (by FISH) is a good predictor for TKI response.

For what does the KRAS mutation or ERCC1 expression predict?

The KRAS mutation or ERCC1 expression predicts for resistance to platinum-based chemo.

image Workup/Staging

What is the initial workup for a pt suspected of having lung cancer?

Lung cancer initial workup: H&P + focus on weight loss >5% over prior 3 mos, Karnofsky performance status (KPS), tobacco Hx, neck exam for N3 Dz, CBC, CMP, CT chest to include adrenals or PET/CT, MRI for paraspinal/sup sulcus tumors, Dx of lung cancer rendered by Bx via transbronchial endoscopic or transthoracic FNA, MRI brain for presumed stages II–III, mediastinoscopy or endobronchial ultrasound (EBUS) for suspected hilar or N2 nodes, PFTs prior to Tx, and smoking cessation counseling

What are the 3 most common presenting Sx of NSCLC?

Dyspnea, cough, and weight loss (others include chest pain and hemoptysis)

What is the sensitivity and specificity of sputum cytology for Dx of lung cancer?

Sensitivity <70%, specificity >90%. Accuracy increases with increasing # of specimens analyzed. At least 3 sputum specimens are recommended for the best accuracy.

What is the sensitivity and specificity of FDG-PET compared to CT for the staging of lung cancers?

1.     PET: sensitivity 83%, specificity 91%

2.     CT: sensitivity 64%, specificity 74%

What is the estimated % of pts who will have false+ N2 nodes based on PET/CT?

~10%–20%. PPV ~80%. +N2 nodes by PET/CT need pathologic confirmation before deferring to potentially curative surgery.

What is the estimated % of pts who will have false+ N2 nodes based on PET/CT?

~5%–16%. NPV ~95%. −N2 nodes by PET/CT for clinical T1 lesions may not need mediastinoscopic evaluation (this is controversial).

What is the rate of occult mets from lung cancer detected by FDG-PET?

In many series, the range is ~6%–18%.

If a PET scan is being ordered, should a bone scan be obtained to evaluate for bone mets as well?

No. In NSCLC, PET is just as sensitive as bone scan but more specific. However, consider pathologic confirmation for solitary PET+ lesions given the risk of a false+.

What clinical characteristics are important to focus on to determine the nature of a solitary pulmonary nodule?

Nodule size (and whether there are changes in size in the past 2 yrs), Hx of smoking, age, and nodule margin on CT (i.e., spiculation)

Stage for stage, which histology has a worse prognosis: adenocarcinoma or SCC? Why?

Adenocarcinoma. It has a greater propensity to metastasize, particularly to the brain.

Does large cell carcinoma have a natural Hx and prognosis more similar to SCC or adenocarcinoma?

Large cell carcinoma has a natural Hx and prognosis more similar to adenocarcinoma.

Describe the T staging of NSCLC using the AJCC 7th edition (2009) of the TNM stage.

1.     T1: ≤3 cm, surrounded by lung parenchyma (T1a ≤2 cm; T1b 2.1–3 cm)

2.     T2: >3–7 cm, +visceral pleura, >2 cm from carina, +atelectasis to lobe (T2a 3.1–5 cm; T2b 5.1–7 cm)

3.     T3: >7 cm, tumor invading mainstem bronchus <2 cm from carina; invasion to diaphragm, chest wall (CW), pericardium, mediastinal pleura; or associated atelectasis or obstructive pneumonitis of entire lung or satellite nodule in same lobe

4.     T4: any size, invading mediastinum, heart, great vessels, trachea, recurrent laryngeal nerve, esophagus, vertebral body, carina, or with separate tumor nodules in a different ipsi lung

Describe the N staging of NSCLC.

1.     N1: ipsi hilar or peribronchial nodes

2.     N2: ipsi mediastinal or subcarinal nodes

3.     N3: any supraclavicular/scalene nodes or contralat mediastinal/hilar nodes

What is the AJCC 7th edition (2009) of the TNM stage for malignant pleural/pericardial nodules/effusion or opposite lung tumor nodules in NSCLC?

M1a, whereas DM is M1b.

What is considered early-stage NSCLC? Categorize the appropriate TNM stratification.

Stages I and II are considered early-stage NSCLC.

1.     Stage IA: T1aN0, T1bN0

2.     Stage IB: T2aN0

3.     Stage IIA: T2bN0 or T1-2aN1

4.     Stage IIB: T2bN1 or T3N0

What procedures prior to thoracotomy can be used to evaluate the following nodal stations: (a) left and right stations 2, 4, and 7; (b) stations 5–6?

1.     Mediastinoscopy to evaluate left and right stations 2, 4, and 7 or EBUS to evaluate left and right stations 2, 3, 4, 7, and 10

2.     Video-assisted thoracic surgery or ant mediastinotomy (Chamberlain procedure) for stations 5–6

When should pre-Tx mediastinal nodal assessment be done?

1.     To confirm PET or CT +nodes

2.     All superior sulcus tumors

3.     If T3 or central T1-T2 lesions

What routine PFT results (FEV1 and DLCO) indicate that the pt needs further testing prior to undergoing resection?

If the FEV1 is <80% predicted for the age and size of the pt or the DLCO is <80% predicted, then the pt may need quantitative lung scans/exercise testing to carefully predict postop pulmonary function.

What is the min absolute FEV1 necessary for pneumonectomy and lobectomy?

1.     Pneumonectomy: >2 L

2.     Lobectomy: >1.5 L

Any <1.5-L pt may be a candidate for wedge resection. The marginal % FEV1 for surgery is 40% of the predicted value.

Which subsets of lung cancer pts are at high risk for surgical morbidity?

Subsets at high risk for surgical morbidity:

1.     pCO2 <45 mm Hg

2.     pO2 <50 mm Hg

3.     Preop FEV1 <40% of predicted value

4.     Poor exercise tolerance

5.     DLCO <50% of predicted value

6.     Postop FEV1 <0.71 or <30% of predicted value

7.     Cardiac problems (left ventricle ejection fraction <40%, myocardial infarction within 6 mos, arrhythmias)

8.     Obesity

What are some factors that predict for postop complications (i.e., mortality, infection)?

Active smoking (6 times higher), poor nutrition, advanced age, and poor lung function. It is advised that pts should quit smoking for at least 4 wks prior to resection and have a nutrition evaluation.

What % of lung cancer pts clinically at stage I are upstaged at surgery?

~5%–25% of stage I lung cancer pts are upstaged at surgery.

In addition to stage, name 3 other poor prognostic factors in lung cancer pts.

Poor prognostic factors in lung cancer:

1.     KPS <80%

2.     Weight loss >5% in 3 mos

3.     Age >60 yrs

image Treatment/Prognosis

Generally, what is a Tx paradigm for a stage I–II medically operable NSCLC pt?

Stage I–II medically operable NSCLC Tx paradigm: surgical resection (lobectomy) + mediastinal LND → adj chemo for stage II

Generally, what is a Tx paradigm for a stage I–II medically inoperable NSCLC pt?

Stage I–II medically inoperable NSCLC Tx paradigm: if T1-2N0, consider definitive hypofractionated stereotactic body radiation therapy (SBRT). Otherwise, use definitive conventional RT alone.

Name 3 surgical options to resect a T1-T2 tumor.

Surgical options to resect a T1-T2 tumor:

1.     Wedge or segmental resection

2.     Lobectomy

3.     Pneumonectomy

For a T1N0 NSCLC, what is the estimated LC for wedge/segmental resection vs. lobectomy?

Wedge/segmental LC is 82%. Lobectomy LC is 94% (LF 18% vs. 6%) based on RCT LCSG 921 (Ginsberg RJ et al., NEJM 1995). Lobectomy is preferred when feasible.

For a T1N0 NSCLC, what is the estimated LC for wedge/segmental resection +/+ intraop brachytherapy?

LC is 97% with brachytherapy compared to 83% without brachytherapy. The outcomes may be equivalent to lobectomy. (Fernando HC et al., J Thorac Cardiovasc Surg 2005)

What % of stage I NSCLC pts will develop a 2nd primary after definitive surgical resection?

Up to 30% of pts develop a 2nd primary.

What is the estimated 5-yr OS of completely resected T1-2N0 NSCLC with no adj chemo?

T1N0 ~80%; T2N0 ~68% (Martini N et al., J Thorac Cardiovasc Surg 1999)

What is the 5-yr OS, CSS, and MS for pts who refuse any Tx for T1-2N0 NSCLC?

5-yr OS is 6%, CSS is 22%, and MS is 13 mos (Raz DJ et al., Chest 2007)

What are the indications for adj chemo after definitive resection for stage I–II NSCLC?

Indications for adj chemo after definitive resection for stage I–II NSCLC:

1.     Stage II–IIIA Dz after resection (category 1)

2.     N1 Dz (category 1)

3.     T2N0, especially if the tumor is >4 cm as per unplanned analysis of CALGB 9633 (category 2b)

What is the estimated 5-yr OS benefit with adj chemo for pts with completely resected stage I or II NSCLC?

~5% at 5 yrs based on LACE meta-analysis of recent trials (Pignon JP et al., JCO 2008)

Is there a role for preop chemo in early-stage lung cancer pts?

No. Based on a meta-analysis, the survival gain is the same as for adj chemo (5%). The largest randomized trial (MRC LU22/EORTC 08012) for 519 pts randomized to preop chemo vs. surgery alone found good RR (49%) and downstaging (31%) but no survival benefit to preop chemo. (Gilligan D et al., Lancet 2007)

Is there a benefit of full mediastinal dissection vs. nodal sampling in early-stage pts undergoing surgical resection?

Possibly. A recent pooled analysis of 3 trials demonstrated a 4-yr OS benefit in stage I–IIIA NSCLC pts (HR 0.78). Mediastinal dissection involves removal of the right 2R, 4R, 7, 8R, and 9R and the left 5, 6, 7, 8L, and 9L.

What are the indications for postop mediastinal RT (PORT) after definitive resection for stage I–II NSCLC?

+Margins, +ECE, and unexpected N2 Dz. Per NCCN 2010, give concurrent CRT for a +margin and sequential CRT for N2 Dz (chemo → RT). Although not specified on NCCN, chemo → RT will also be recommended for ECE.

What RT dose is used for a +margin after surgery?

1.     Microscopic + margin: 54–60 Gy

2.     Gross + margin: 60–70 Gy

What randomized study demonstrated improved LC and survival with PORT after surgical resection for early-stage (stages Ia and Ib) NSCLC?

Italian study (Trodella L et al., Radiother Oncol 2002): 104 pts, stage I, resected with LND (pN0), randomized to PORT vs. observation; PORT encompassed the bronchial stump + ipsi hilum (mean area, 50 cm2) to 50.4 Gy. There were better LF rates (2% vs. 23%) with a trend to improved survival (67% vs. 58%, p = 0.048). There was min toxicity and no worsened pulmonary function.

What is the max RT dose that has been used for definitive RT alone in stage I–II NSCLC?

Up to 84 Gy in standard fractionation if lung dose volume constraints are respected

What is the 2-yr local recurrence rate for RT alone using standard fractionation?

50%–78% 2-yr local recurrence is based on RTOG 9311, but this study included stage III pts as well.

In stage I–II NSCLC pts being treated with definitive RT alone, should elective nodal regions be treated? What is the estimated elective nodal failure rate if untreated?

No. Elective nodal failure rate was <10% in RTOG 9311. In most series with stage I lung cancer treated with hypofractionated SBRT, the regional nodal failure rate ranged from 5%–10%. From the Indiana University dose escalation study using SBRT for stage I lung cancer, regional nodal failure as the site of 1st failure was 10%. (Hoopes DJ et al., Lung Cancer 2007)

What is the estimated 3- to 5-yr LC and OS for medically inoperable stage I pts treated with definitive hypofractionated SBRT?

In most series using hypofractionated SBRT for stage I lung cancer pts, the 3-yr LC ranged from 85%–95% and the 3-yr OS was 55%–91%.

What biologic effective dose (BED) should be achieved to attain max LC and survival in pts with stage I lung cancer treated with SBRT?

If the BED is ≥100 Gy, then the 5-yr LC rate is 92% and the 5-yr OS is 72%. However, if the BED is <100 Gy, then the 5-yr LC rate is 57% and the 5-yr OS is 50%. (Onishi H et al., JTO 2007)

What is the 5-yr rate of DM in early-stage NSCLC after definitive SBRT?

~15%–25%, which is very similar to the rate seen after surgical resection

What is the SBRT technique that was evaluated in RTOG 0236? What is the 2-yr LC and OS rate for this group of inoperable pts?

SBRT using 20 Gy × 3 (without homogeneity correction) given in 1.5–2 wks. Dose was 18 Gy × 3 with homogeneity correction. 3-yr LC was 90.6%, DFS was 48.3%, and OS was 55.8%. (Timmerman R et al., JAMA 2010)

Are centrally located or peripherally located lesions a relative contraindication for hypofractionated SBRT?

Lesions located centrally, within 2 cm of the central bronchial tree, are not good hypofractionated RT candidates using 20 Gy × 3 fx due to the risk of grade 3–5 toxicity (Timmerman R et al., JCO 2006). MDACC (Chang JY et al., IJROBP 2008) proposed 12.5 Gy × 4, with LC at 17 mos of 100%. There was grade 2–3 dermatitis and CW pain in 11%. There was no pneumonitis for newly diagnosed stage I pts.

At the Johns Hopkins Hospital, 4 Gy × 15 or 5 Gy × 12 are used for central lesions.

What studies are currently addressing the role of SBRT in resectable early-stage lung cancer pts?

1.     RTOG 0618: phase II, operable stage I–II NSCLC, accrual goal of 33 pts using 18 Gy × 3 fx.

2.     ROSEL (Dutch): phase III SBRT vs. surgery for operable stage I pts. RT 20 Gy × 3 for T1, 12 Gy × 5 for T2, and 7.5 Gy × 8 for central tumors. Primary endpoint is 2- and 5-yr LC, QOL, and cost.

3.     STARS (Accuray/MD Anderson): phase III SBRT (CyberKnife) vs. surgery in operable stage I pts. RT 12.5 Gy × 4 for central lesions and 16.7 Gy × 3 fx for peripheral tumors. Primary endpoint is OS, DFS, and toxicity at 3 yrs.

image Toxicity

What is the typical follow-up schedule of pts treated for lung cancer?

Typical lung cancer follow-up: H&P + CT chest with contrast q4–6mos for yrs 1–2, then noncontrast CT chest annually for yrs 3–5; continued smoking cessation counseling (NCCN 2010)

What are the toxicities seen with SBRT for early-stage lung cancer?

Pneumonitis, lung fibrosis/consolidation, cough, dermatitis, CW pain, esophagitis, and hemoptysis

What is the total lung V20 dose-volume constraint for RT alone?

The total lung V20 dose-volume constraint is <37%. (NCCN 2010)

What is the mean lung dose (MLD) constraint for definitive RT to lung cancer?

The MLD constraint is <20 Gy. (NCCN 2010)

What is the distinction between grade 2 and 3 RTOG pneumonitis?

1.     Grade 2 pneumonitis: symptomatic with the need for steroids

2.     Grade 3 pneumonitis: dyspnea at rest and oxygen supplementation needed

3.     Other grades of pneumonitis include grade 1: asymptomatic, seen only on CT; grade 4: hospitalized and intubated; grade 5: death

What are the heart dose-volume constraints for conventionally fractionated RT?

Heart V45 <67% and V60 <33%. (NCCN 2010)

What is the dose constraint for the brachial plexus with conventional fractionation?

The dose to the brachial plexus should be kept at <66 Gy.

What is the rate of brachial plexopathy seen for pts treated with SBRT for early-stage lung cancer?

The recommended max dose to the brachial plexus is <26 Gy in 3–4 fx (Forquer JA et al., Radiother Oncol 2009). Of the 37 apical tumors out of 253 pts treated, 19% of pts developed grade 2–4 plexopathy. A dose >26 Gy resulted in 46% risk vs. 8% risk if the dose was <26 Gy.

What is the esophageal dose constraint for conventionally fractionated RT?

Ideally, the mean dose of RT to the esophagus should be <34 Gy. Try to minimize the V60 as much as possible (V60 <33%, V55 <66%, ≤45 Gy to the entire esophagus).

What is the max BED for SBRT when treating centrally located tumors?

180–210 Gy with grade 3 pulmonary complications (Timmerman R et al., JCO 2006). Keeping the BED ≥100 Gy may be sufficient for LC and may avert toxicities for central lesions.

Is a normal SUVmax of FDG-PET required to be considered a good clinical response in the follow-up of stage I NSCLC pts treated with SBRT?

No. Prospective and retrospective reviews suggest that the SUVmax remains elevated for an extended period after SBRT due to an inflammatory response, but there is no evidence of correlation with recurrence. (Timmerman R et al., IJROBP 2009; Henderson MA et al., IJROBP 2009)

What % of pts treated with SBRT for early-stage lung cancer have grade 3–5 toxicities?

1.     15% of pts have grade 3–5 toxicities in a review of 15 studies (683 pts). (Sampson JH et al., Semin Radiat Oncol 2006)

2.     In RTOG 0236, 16.3% of pts experienced grade 3–4 toxicity but no grade 5 toxicity. (Timmerman R et al., JAMA 2010)

What factors predict for grade 3–5 toxicities after SBRT for early lung cancer seen on the Indiana University phase II study?

Location (46% hilar/pericentral vs. 17% peripheral) and tumor size (GTV >10 cc had 8 times the risk of grade 3–5 toxicity.) (Timmerman R et al., JCO 2006)

What are the PFT changes before and after SBRT for early-stage lung cancer?

Very min based on an institutional review of 92 pts (Stephans KL et al., JTO 2009): mean FEV1 −0.05 (−1.88%), DLCO −2.59% of predicted value; no association with central vs. peripheral location or the dose administered