Radiation Oncology: A Question-Based Review

Small Cell Lung Cancer and Bronchial Neuroendocrine Tumor

Boris Hristov and Ritsuko Komaki

image Background

Small cell lung cancer (SCLC) accounts for what % of new lung cancer Dx in the U.S.? What % of lung cancer deaths?

~15% (~33,000 cases/yr) of new lung cancer is SCLC, accounting for 25% of lung cancer deaths annually.

What % of SCLC is linked to smoking?

>90% of SCLC cases are linked to smoking.

What is the median age of Dx of SCLC? What % of pts are >70 yrs old at Dx?

The median age of SCLC Dx is 64 yrs, with 25% of pts presenting at age >70 yrs.

What % of pts with SCLC presents with metastatic Dz?

67% of SCLC pts present with mets, most commonly to the contralat lung, contralat or bilat malignant pleural effusion, liver, renal, adrenals, bone, BM, and brain.

What are the pathologic characteristics of SCLC?

Small round blue cells of epithelial origin with neuroendocrine differentiation, ↑ mitotic count, and ↑ N/C ratio

What are the markers that characterize SCLC?

Markers that characterize SCLC include S100, synaptophysin+, chromagranin+, and neurotensin + EGFR−.

What are some common neurologic and endocrine paraneoplastic syndromes associated with SCLC?

1.     Neurologic: Lambert-Eaton syndrome (antibody to presynaptic voltage-gated calcium channels), encephalomyelitis, sensory neuropathy (anti-Hu antibody)

2.     Endocrine: Cushing Dz (↑ ↑ ACTH), syndrome of inappropriate secretion of antidiuretic hormone (SIADH) (↑ ↑ADH)

What is the most common chromosomal abnormality associated with SCLC but not seen with extrapulmonary small cell carcinomas?

Deletion of 3p (95% of cases, particularly 3p14-25 region, with inactivation of at least 3 tumor suppressor genes, including FHIT and RASSF1A)

What is the most common genetic alteration seen in SCLC?

Amplification of the bcl-2/C-myc family of oncogenes is most common but likely is not the initiating event. Other common abnormalities include loss of p16, loss of Rb, and mutation in p53.

image Workup/Staging

How do pts with SCLC usually present?

Large hilar mass with bulky mediastinal LAD that causes cough, shortness of breath, weight loss, postobstructive pneumonia, and debility. Other common presentations include paraneoplastic syndromes such as Lambert-Eaton, SIADH, or ectopic ACTH production.

Classically, does SCLC present centrally or peripherally in the lung?

Classically, SCLC presents centrally in the lung.

What histology is most commonly associated with superior vena cava obstruction (SVCO) syndrome?

SCLC is most commonly associated with SVCO syndrome.

Do SCLC pts present with solitary peripheral nodules without mediastinal LAD?

This is very uncommon (<3%).

How should pts be managed whose FNA results cannot clearly differentiate between small cell and atypical carcinoid histology?

Surgical staging, with mediastinoscopy → surgical resection if the mediastinum (MN) is negative (per the NCCN)

Once SCLC has been diagnosed in a pt who presents with a large hilar mass, what further workup is necessary besides the basic H&P and labs?

LDH levels, CT C/A/P +/− PET, MRI brain, bone scan if PET is not done, BM Bx (for pts with elevated LDH), thoracentesis with cytopathologic exam for pts with pleural effusion, and smoking cessation counseling

What % of pts with SCLC at the time of Dx present with brain mets, BM involvement, and bone mets?

1.     Brain mets: 10%–15% (30% are asymptomatic)

2.     BM involvement: 5%–10%

3.     Bone mets: 30%

What is the latest AJCC system for staging SCLC?

The same as for non-SCLC, but this system is not commonly used.

How is SCLC most commonly staged?

SCLC is staged using the International Association of Lung Cancer system, which is a modification of the Veterans Administration Lung Cancer Study Group (VALCSG) system. There are 2 stages: limited and extensive. Tumors are staged according to whether the Dz can be encompassed within an RT port. Limited stage Dz is typically confined to the ipsi hemithorax, without malignant pleural effusion, contralat Dz, or mets; other presentations are usually extensive stage.

What % of pts present with limited-stage SCLC?

~33% of pts present with limited-stage SCLC.

What are some adverse prognostic factors in SCLC?

Poor performance status (PS); weight loss (>5% in prior 6 mos); ↑LDH; male gender; endocrine paraneoplastic syndromes (controversial), variant, or of mixed cell type; metastatic Dz

What is the MS of untreated limited-stage and extensive-stage SCLC?

~12 wks for limited stage and ~6 wks for extensive stage, based on a VALCSG trial comparing cyclophosphamide to placebo.

What is the MS for pts with limited- vs. extensive-stage SCLC?

1.     Limited stage: 19–23 mos (Turrisi AT et al., NEJM 1999)

2.     Extensive stage: 5–7 mos (Slotman B et al., NEJM 2007)

What is the long-term survival rate in limited-stage SCLC treated with a combined modality?

~20%–30% long-term survival (5-yr OS)

What additional workup should be considered for pts with carcinoid tumors of the lung?

Consider octreotide scan.

image Treatment/Prognosis

What is the Tx paradigm for pts with extensive-stage SCLC?

Extensive-stage SCLC Tx paradigm: multiagent chemo regimen including etoposide/cisplatin (EP) or Cytoxan/Adriamycin/vincristine (CAV). Consider consolidation RT to the thorax for pts who achieve a CR to distant Dz after initial chemo. (Jeremic B et al., JCO 1999)

What are some important poor prognostic factors for limited stage SCLC?

Karnofsky PS <70, weight loss, ↑LDH

What is the Tx paradigm for pts with limited-stage SCLC?

Limited-stage SCLC Tx paradigm: 4 cycles of EP chemo (etoposide [120 mg/m2, days 1–3] + cisplatin [60 mg/m2, day 1, q3wks]) + concurrent RT (only 1 cycle is concurrent). Current standard RT regimen is based on INT-0096: 45 Gy in 1.5 Gy bid × 30 fx.

How should a tumor characterized as a high-grade neuroendocrine carcinoma, or as large cell neuroendocrine carcinoma, be managed?

Treat per non-SCLC guidelines. (NCCN 2010)

How should a pt with a carcinoid tumor of the lung be managed?

Surgery is 1st line if it is resectable. Some centers will treat atypical carcinoid per the SCLC paradigm (nonsurgical).

Per NCCN 2010:

1.     Stage I–III typical carcinoid tumor can be observed after R0 resection.

2.     For unresectable or medically inoperable stage III typical carcinoid tumors, RT alone is recommended.

3.     For atypical carcinoid tumors, resected stage I can be observed. However, for resected stage II–IIIa tumors, chemo (EP) and RT adjuvantly is recommended.

How should stage IIIB–IV or unresectable carcinoid of the lung be managed?

Systemic therapy (EP), or octreotide if octreotide scan positive or symptomatic from paraneoplastic syndrome (NCCN 2010)

What is the benefit of smoking cessation prior to Tx in pts with limited-stage SCLC?

Toxicity and ↑survival, based on a retrospective review (Videtic GMM et al., IJROBP 2003)

What are the typical response rates seen after concurrent CRT for limited-stage SCLC?

Typical response rates are 80%–95% with CR rates of 40%–60%.

What is the median duration of response for pts with SCLC after definitive Tx?

6–8 mos is the median duration of response.

What is the preferred Tx approach for elderly pts (age >70 yrs) with limited-stage SCLC?

Depends on PS. In pts with good PS, combined CRT is preferred. Otherwise, standard combination chemo is better than single-agent cytotoxic agents.

What is the MS of SCLC pts after recurrence if treated with salvage chemo?

4–5 mos is the MS for these pts.

Why is EP the preferred regimen in concurrent CRT for limited-stage SCLC?

EP causes little mucosal toxicity, offers low risk of interstitial pneumonitis, and has lower cardiac toxicity compared to Adr. Full systemic doses can be administered with RT, and there is modest hematologic toxicity. EP has a better therapeutic ratio over the older regimen of CAV, but it confers no survival benefit.

What are the benefits and disadvantages of substituting carboplatin for cisplatin in EP for Tx of SCLC?

Carboplatin is less emetic/neuropathic/nephropathic/ototoxic, but there is more heme toxicity.

Is there a benefit to maintenance chemo after the initial 4–6 cycles in the Tx of SCLC?

No. Maintenance chemo only produces minor prolongation of response without improving survival and increasing cumulative toxicity.

Is there a benefit of irinotecan compared to etoposide when added to cisplatin in the Tx of SCLC?

A Japanese RCT demonstrated a survival benefit with irinotecan, but this was not reproduced in a U.S. trial that used a modified (lower) dose from that used in the Japanese regimen. Now, a new phase III U.S. trial is testing the exact Japanese regimen.

What is the OS and LC benefit of adding RT to chemo in limited-stage SCLC?

There is an OS benefit of 5% based on Pignon meta-analysis (1992), with LC benefit of 25%–30% (Warde & Payne meta-analysis: NEJM 1992).

What is the optimal sequence of combining chemo with RT?

Concurrent is better than sequential (JCOG: Takada M et al., JCO 2002): MS 27 mos vs. 20 mos; 5-yr OS 30% vs. 20% (10% OS benefit)

What evidence supports early concurrent CRT → chemo over late RT with induction CT → CRT?

NCIC data (Murray N et al., JCO 1993): phase III, 308 pts. 5-yr OS was 20% (early RT) vs. 11% (late RT).

Yugoslavia data (Jeremic B et al., JCO 1997): an early vs. late trial showed better MS (34 mos vs. 26 mos) and 5-yr OS (30% vs. 15%) for early.

Meta-analysis of 7 trials (Fried DB et al., JCO 2004): early (<9 wks) vs. late (>9 wks) after chemo. There was 5.2% better 2-yr OS with early RT.

What is the recommended RT dose in CRT for limited-stage SCLC?

45 Gy in 1.5 Gy bid or 60–70 Gy in 2 Gy qd (NCCN 2010) or 61.2 Gy in 5 wks (1.8 Gy qd × 16 fx → bid as in RTOG 97-12 (Komaki R et al., IJROBP 2005)

What randomized trial demonstrated a clear superiority of altered fractionation with chemo compared to qd RT in the Tx of SCLC?

INT-0096 (Turrisi AT et al., NEJM 1999): phase III, 381 pts, EP × 4 cycles + RT at 1st cycle; randomization with 1.5 Gy bid × 3 wks vs. 1.8 Gy qd × 5 wks (both to 45 Gy); all rcv PCI to 25 Gy. There was better 5-yr OS (26% vs. 16%) and LC (64% vs. 48%) in the bid arm. There was increased grade 3 esophagitis (27% vs. 11%) in the bid regimen, but not in the grade 4 toxicity. Criticism: 45 Gy qd is not biologically equivalent to the accelerated hyperfractionation of 45 Gy in 30 fx.

What studies support dose escalation with conventional fractionation over the traditional bid approach?

CALGB 8837 (Choi H et al., JCO 1998): phase I MTD in 50 pts of 2 RT regimens: 1.5 Gy/fx bid or 2.0 Gy/fx qd; MTD of bid was 45 Gy, whereas MTD of qd regimen was >70 Gy. Updated survival results were that 6-yr OS was better in the qd regimen compared with bid (36% vs. 20%). A phase III trial is ongoing (CALGB 30610/RTOG 0538) to compare 45 Gy in 3 wks (arm A: INT-0096) vs. 70 Gy in 35 fx (arm B: CALGB regimen) vs. 61.2 Gy in 5 wks (arm C: MTD in dose escalation study RTOG 97-12 also used in RTOG 0239).

Is there evidence to support consolidative RT to thoracic Dz in extensive-stage SCLC?

Yes. Jeremic et al. enrolled 210 extensive-stage pts. All rcv EP × 3. The subset of 109 pts who achieved a CR at all distant sites was randomized to rcv consolidative RT (accelerated hyperfractionation to 54 Gy) with concurrent carboplatin/etoposide or not. In both arms, pts rcv PCI and consolidative EP. Consolidative CRT improved 5-yr OS (9.1% vs. 3.7%) and MS (17 mos vs. 11 mos). There was a trend in favor of LC but not DM-free survival. (JCO 1999)

Is there a role for definitive surgical resection in SCLC pts?

This is controversial and not standard. Retrospective studies suggest that with modern staging, T1-2N0 SCLC pts have reasonable outcomes with surgery and adj platinum-based chemo. The situation is uncommon (~5% of cases). (Brock MV et al., J Thorac Cardiovasc Surg 2005)

A JCOG phase II study tested surgical resection for stage I–IIIA SCLC → EP chemo. 5-yr OS was 69%, 38%, and 40%, respectively. Currently, RCTs in Europe and Japan are testing surgical management.

What is the recommended adj Tx for SCLC if the mediastinal nodes are found to be involved after attempted surgical resection?

Concurrent CRT directed at the MN (NCCN 2010); if node–, adj chemo alone

What additional chemo agents, when added to EP, have been shown to modestly improve the survival of pts with extensive-stage SCLC?

Ifosfamide or cyclophosphamide + an anthracycline have been shown to modestly improve survival.

What are some salvage chemo agents used at the time of recurrence for SCLC?

Oral topoiseromase I inhibitors are standard for post-chemo failure. (Topotecan was tested in RCTs showing doubling of survival [26 wks vs. 14 wks] compared with supportive care; irinotecan as a single agent was not tested.) The old combo chemo of CAV can be used (if EP is used as 1st line). Paclitaxel, docetaxel, vinorelbine, and gemcitabine as single agents are not standard since activity is low against SCLC.

What is the role of PCI in limited-stage SCLC? Is there an OS benefit with it?

Auperin meta-analysis of 7 RCTs (NEJM 1999) compared PCI vs. no PCI after CR following induction chemo +/− RT and no evidence of brain mets before randomization. There was ↓ 3-yr incidence of brain mets (33% vs. 59%) and 5.4% better 3-yr OS (20.7% vs. 15.3%) and improved DFS. There was a trend to a better outcome with ↑ doses and RT <4 mos from the start of chemo.

What PCI dose is now standard for limited-stage SCLC?

25 Gy in 10 fx is now the standard dose for PCI for limited-stage SCLC. (RTOG 0212-Intergroup: Le Pechoux C et al., Lancet Oncol 2009)

For pts with limited-stage SCLC, what PCI doses were compared in RTOG 0212?

Standard doses (25 Gy in 10 fx) vs. higher doses (36 Gy in either 18 fx qd or 24 fx bid). There was no difference in the 2-yr incidence of brain mets, but there was an OS and chest relapse advantage for the standard arm (42% vs. 37%, p = 0.05) due to greater cancer-related mortality in the high-dose group. (Le Pechoux C et al., Lancet Oncol 2009)

What is the effect of PCI timing after the initiation of chemo for SCLC?

Based on Auperin meta-analysis, there was a decrease in risk of brain mets with earlier PCI (<4–6 mos vs. >6 mos) without an effect on risk of death.

Is there data demon-strating greater neuro-psychologic complications after PCI for SCLC?

No. The data actually demonstrate no difference with or without PCI in a randomized trial addressing the question of neuropsychologic changes after PCI (Arrigada et al., JNCI 1995). Most pts (97%) actually have abnl neuropsychologic testing after chemo and before PCI, without a difference after PCI (Komaki R et al., IJROBP 1995).

What seminal study demonstrated a survival benefit with PCI in pts with extensive-stage SCLC with any response to chemo? What are some main criticisms of this study?

EORTC 08993 RCT (Slotman BJ et al., NEJM 2007): 286 pts with extensive-stage SCLC treated with chemo; primary endpoint was time to symptomatic brain mets; pts randomized to +/− PCI after any response to chemo; most PCI pts given 20 Gy in 5 fx. PCI lowered the risk of symptomatic mets and improved DFS and OS (5.4 mos –PCI vs. 6.7 mos +PCI). 1-yr OS nearly doubled (13% –PCI vs. 27% +PCI).

Criticisms: There was no pre-Tx MRI. The RT group was more likely to rcv chemo at the time of extracranial progression (68% vs. 45%). Only about half (59%) of pts in the control group rcv WBRT for intracranial progression of Dz.

What was the greatest QOL alteration after PCI in the EORTC trial for extensive-stage SCLC?

3-mo QOL assessment showed that the largest negative impact of PCI was fatigue and hair loss. Worsening role, emotional, and cognitive function were also seen after PCI. (Slotman BJ et al., JCO 2009)

What is the current recommendation for PCI in pts with SCLC?

Limited or extensive stage, CR/PR after chemo +/− RT, +/− MRI brain, PS of ECOG 0–2, within 3–6 wks of last cycle of chemo, 25 Gy in 10 fx. In pts with less than a CR, PCI is at the discretion of the treating physician.

What chest RT volumes are used to treat limited-stage SCLC?

1.     Classic fields: gross tumor, ipsi hilum, bilat MN from thoracic inlet (1st rib) down to 5 cm below the carina. CTV = GTV + 1.5 cm (and elective hilum and MN regions + 8 mm), PTV = CTV + 1 cm.

2.     CALGB 30610/RTOG 05383D-CRT or IMRT allowed. CTV needs to include GTV plus potential occult Dz defined as (a) ipsi hilum (level 10 LN) and (b) N2-N3 levels from the top of the aortic arch down to 3 cm below the carina encompassing levels 3, 4R, 4L, and 7 (and levels 5–6 for left-sided tumors). These areas are treated electively except for the supraclavicular fossa. PTV is CTV + 0.5 cm if daily setup imaging is used and if ITV assessment is done during simulation and the planning process (either breath-hold or 4D-CT imaging). If a free-breathing non-ITV approach is used (non–4D-CT simulation), the PTV is CTV + 1.5 cm (sup-inf direction) and 1.0 cm in the axial direction. If a breath-holding non-ITV, PTV is CTV + 1.0 cm (sup-inf direction) and 0.5 cm in the axial direction.

What thoracic RT field arrangements are used for treating limited-stage SCLC?

IMRT if V20 <30% and FEV1 >1 L. Otherwise, minimize contralat lung exposure with AP/PA to 15 Gy (1.5 Gy bid × 5 days) with oblique field CD to 45 Gy (AP/PA in am and obliques for pm sessions).

In SCLC pts with SVCO, cord compression, or brain mets, what regimen is preferred as upfront palliative Tx: RT or chemo?

In a chemo-naive pt presenting with SVCO, RCTs have shown a similar symptomatic response rate with chemo compared with RT. But in a chemo-refractory pt, RT is the preferred regimen. In pts with cord compression/brain mets, RT is standard (in both chemo-naive and chemo-refractory pts).

How is palliative RT delivered for SVCO syndrome in pts with SCLC?

Generally, a few large fx upfront (3–4 Gy × 2–3) → more definitive dosing in conventional fractionation (qd or bid regimen)

What fractionation scheme is optimal for pts with lung cancers treated with palliative RT for Sx such as hemoptysis, cough, pain, and shortness of breath?

Conventional is probably no better than hypo-fractionation. In a Norwegian RCT, Sundstrom et al. tested 30 Gy in 10 fx vs. 17 Gy in 2 fx (1 wk apart) vs. 10 Gy for 1 fx. All achieved similar levels of palliation. (JCO 2007)

image Toxicity

What is the recommended follow-up schedule for SCLC pts?

SCLC follow-up schedule: H&P, CT chest, and labs at each visit (visits q2–3mos for yr 1, q3–4mos for yrs 2–3, q4–6mos yrs 4–5, then annually). PET scan should be considered whenever CT findings suggest recurrence or mets.

What is the total lung V20 dose-volume constraint for RT alone and concurrent CRT in definitive lung cancer Tx?

1.     RT alone: V20 <40%

2.     CRT: V20 <35%

(NCCN 2010)

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

MLD is <15 Gy ideally but not >20 Gy.

What is the max cord dose allowed with the Turrisi bid regimen?

With the Turrisi bid regimen, the max cord dose is 36 Gy.

What is the main toxicity associated with using bid RT as done in the Turrisi regimen?

Grade 3 acute esophagitis: 27% (bid) vs. 11% (qd). Other toxicities (myelosuppression, nausea) were the same as the qd regimen.

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

1.     Grade 3 pneumonitis: dyspnea at rest or oxygen supplementation needed

2.     Grade 2 pneumonitis: symptomatic and not requiring oxygenation

What is the heart dose-volume constraint for RT alone vs. concurrent CRT?

1.     RT alone: V40 <50%

2.     CRT: V40 <40%

According to RTOG 0538, the following limits are also acceptable: 60 Gy less than one third, 45 Gy less than two thirds, and 45 Gy <100%.

What is the esophageal dose-volume constraint for RT alone vs. concurrent CRT?

1.     RT alone: V60 <50%

2.     CRT: V55 <50% (ideally, keep the mean dose to <34 Gy per RTOG 0538)