The Paradox of the Efficacy of Local Treatment for Small-Cell Lung Cancer

September 1, 2007

Small-cell lung cancer (SCLC) appears to be declining in frequency. This reduced incidence of SCLC seems to have curtailed interest and investigation as well.

Continuing Medical Education InformationThe Paradox of the Efficacy of Local Treatment for Small-Cell Lung Cancer


Activity Release Date: September 1, 2007
Activity Expiration Date: September 1, 2008

About the Activity
This activity is based on a brief article developed as part of the E-Update Series and posted on the Web. The series is geared to oncologists and addresses new treatments of cancer or modifications thereof.

This activity has been developed and approved under the direction of Beam Institute.

Activity Learning Objectives
After reading this article, participants should be able to:

Appreciate the public health problem posed by small-cell lung cancer (SCLC), which despite declining incidence still accounts for 4% of all cancer mortality.Understand the dismal outlook and limited survival for patients with SCLC.Identify sites of metastases and patterns of relapse among SCLC patients.Single out the combination of etoposide and cisplatin as the standard, although not the optimal chemotherapy for SCLC.List the key findings of studies investigating the role of thoracic radiotherapy in SCLC and radiotherapy innovations that have produced survival benefits for SCLC patients.Understand the role of timing when combining radiotherapy and chemotherapy and the importance of combining early thoracic radiation with a compatible chemotherapy regimen.Appreciate the impact of technology that allows more precise targeting of radiotherapy.Know that treating uninvolved nodes with radiotherapy may not be necessary and may increase toxicity risk.Be familiar with variations in radiotherapy dose, time, and fractionation.Identify the established benefits of and lingering concerns about prophylactic cranial irradiation.

Target Audience
This activity targets physicians in the fields of oncology and hematology.

This activity has been planned and implemented in accordance with the Essential Areas and policies of the Accreditation Council for Continuing Medical Education through the joint sponsorship of Beam Institute and The Oncology Group. Beam Institute is accredited by the ACCME to provide continuing medical education for physicians.

Continuing Education CreditAMA PRA Category 1 Credit™
The Beam Institute designates this educational activity for a maximum of 2 AMA PRA Category 1 Credit(s)™. Physicians should only claim credit commensurate with the extent of their participation in the activity.

Compliance Statement
This activity is an independent educational activity under the direction of Beam Institute. The activity was planned and implemented in accordance with the Essential Areas and policies of the ACCME, the Ethical Opinions/Guidelines of the AMA, the FDA, the OIG, and the PhRMA Code on Interactions with Healthcare Professionals, thus assuring the highest degree of independence, fair balance, scientific rigor, and objectivity.

However, Beam Institute, the Grantor, and CMPMedica shall in no way be liable for the currency of information or for any errors, omissions, or inaccuracies in the activity. Discussions concerning drugs, dosages, and procedures may reflect the clinical experience of the author(s) or may be derived from the professional literature or other sources and may suggest uses that are investigational in nature and not approved labeling or indications. Activity participants are encouraged to refer to primary references or full prescribing information resources. The opinions and recommendations presented herein are those of the author(s) and do not necessarily reflect the views of the provider or producer.

Financial Disclosures
Dr. Govindan receives research support from Bristol-Myers Squibb, Eli Lilly, Genentech, Pfizer, Sanofi-Aventis, and serves on the speakers' bureau for Eli Lilly and Genentech.  Dr. Turrisi has no significant financial interest or other relationship with the manufacturers of any products or providers of any service mentioned in this article.

Copyrights owned by Beam Institute, a division of CME LLC. Copyright 2007, CME LLC. All rights reserved.

Contact Information
We would like to hear your comments regarding this or other activities provided by Beam Institute. In addition, suggestions for future activities are welcome. Contact us at:

Director of Continuing Education
Beam Institute
11 West 19th Street, 3rd Floor
New York, NY 10011-4280

Phone: 888-618-5781
Fax: 212-600-3050


Although small-cell lung cancer (SCLC) is a systemic disease all or nearly all of the time, advances in treatment of this systemic disease have come from radiotherapy innovations over the past 20 years. The best results in long-term survival for limited SCLC have been achieved by the addition of thoracic irradiation to systemic chemotherapy, and more intense thoracic irradiation, with early integration of thoracic irradiation and concurrent chemotherapy. The addition of prophylactic cranial irradiation (PCI) adds to survival in complete responders as proven by meta-analytic data. To the surprise of many, a recent trial proves that extensive stage SCLC benefits from PCI after any response to systemic therapy provided for 4 to 6 cycles.

Findings from the American Society of Clinical Oncology as well as a broader review of small-cell lung cancer therapy are summarized in a Journal of Thoracic Oncology overview coauthored with Nevin Murray.1 This E-Update relates the advances in local therapy, particularly radiotherapy, that have made a profound, although often ignored, difference in this systemic disease.

Incidence and Outlook

Small-cell lung cancer (SCLC) appears to be declining in frequency.2 This reduced incidence of SCLC seems to have curtailed interest and investigation as well. The tempo of SCLC investigation does not match the public health problem since SCLC still accounts for 4% of all cancer mortality.3

Untreated SCLC has a dismal outlook, with a median survival of 2 months for extensive small-cell lung cancer (ESCLC) and 3 months for limited small-cell lung cancer (LSCLC).4 Despite high initial responses to chemotherapy, SCLC proves ultimately resistant in most cases. Median survivals of 20 months and 5-year survival of 25% to 30% for patients treated with chemoradiation are very similar for LSCLC and stage III non-small-cell lung cancer (NSCLC).5,6 (See Table 1.)The interval from diagnosis to emergence of resistant clones, and then fatal outcome also seems comparable in SCLC and NSCLC.

Click to enlarge

Although distant metastasis dominate, local and brain relapse remain problems, and disease control appears to plateau with 5-year survival of 20% to 30% in patients treated with combined modalities. While it appears that we are able to remove the visible cells, therapy seems to select for resistant clones. The patient with either limited or extensive SCLC has no chance once a resistant genotypic cell leaves the site of origin and establishes itself and grows distantly.

It may be that the small cells mimic primitive progenitor cells, especially those committed to forming the embryonic lung.7 With stem cells, the ability to migrate or circulate seems critical to their function during organogenesis. Since this also characterizes the metastatic phenotype in general and particularly in lung cancer; we need to identify characteristics and genes related to these functions. We are very able to eliminate the clusters of developed small cells, but the stem cells or progenitors are elusive, and the resistant clones return to doom the patient.

Although a truly effective chemotherapy combination has never become evident, the combination of cisplatin and etoposide is the internationally accepted standard for limited and extensive SCLC, except in Japan where irinotecan (CPT-11, Campath) is used instead of etoposide.8 Somewhat disconcerting and harkening back to the past, the CAV or CAE regimen (cyclophosphamide, doxorubicin [Adriamycin], vincristine and/or etoposide) has persisted in use for more than 30 years, including use in one of the few limited SCLC trials published this century.9

Interest in the combination of etoposide and cisplatin (EP)10 moved forward after this combination produced tumor regression in patients whose cancers had progressed following a cyclophosphamide-based regimen.11 The consistent performance of EP (or acceptably, carboplatin and etoposide) in clinical trials and its compatibility with thoracic radiotherapy allowed both to be used at doses that are known to be systemically active. EP is better than alkylator-based or anthracycline-based regimens,12 and it is possible that alkylator and anthracycline use is detrimental to survival.

Combined Modality Therapy

Thirteen randomized studies, including 2,140 patients, have investigated the role of thoracic radiotherapy in LSCLC. Two meta-analyses13,14 were published 15 years ago. The trials included all used alkylators and/or anthracyclines as their core chemotherapy, but none used the contemporary standard, EP. Nevertheless, both meta-analyses show a modest but consistent improvement in survival rates as well as local control with the addition of thoracic radiotherapy (TRT). (See Table 2.)

Click to enlarge

Survival benefit becomes distinct at about 15 months after the start of treatment and persists beyond 5 years. The 2-year local failure rate of 23% for irradiated patients vs 48% for nonirradiated patients remains significant as well (P = .0001). The treatment-related death rate, however, was only 1%. 

The benefit of radiotherapy in the meta-analyses seemed to be greatest for those younger than 55 years of age (P = .01). However, two more modern studies15,16 show the benefit of TRT in older (≥ 70 yrs) vs younger (<70 yrs) patients administered early EP with thoracic irradiation. A plausible explanation of the discrepancy would be that the toxicity of concurrent chemoradiation with alkylators and anthracyclines, or that the tendency to use longer systemic treatment, as used in the trials examined in the meta-analyses, adversely affected elderly patients. Clearly, fit elderly patients with LSCLC warrant consideration for chemoradiation.

Timing of Combined Therapy

Strict meta-analytic methods cannot prove a benefit of “early” over “late” radiotherapy, according to my own analysis. (A thorough discussion of this can be found in the Journal of Thoracic Oncology concise review mentioned earlier.1) Early thoracic irradiation cannot perform well unless it is coupled with a compatible chemotherapy regimen. The trials that use the EP regimen do show a benefit for early therapy. It is only the trials incorporating an alkylator/anthracycline regimen or trials employing induction cycles of chemotherapy that fail to show a benefit for early therapy. 

The duration of the chemoradiation for LSCLC may be defined as the time elapsed from the first therapeutic intervention until the completion of the radiotherapy treatment. Initial concurrent chemoradiation is the most efficient way to rapidly destroy cancer cells, and eliminate the source of the initial tumor stem cell and the most likely cauldron for emergence of resistant strains. Based on in vitro assays of radiosensitivity of human small-cell lung cancer lines,17 more cancer clonogens are eliminated locally by the first 2 Gy fraction (SF2) than by the entire remainder of the radiotherapy course.

The concept of induction/neoadjuvant chemotherapy is appealing because of instant access of drugs to virgin tumors with intact vasculature and an ability to assess response. However, the biological consequences of induction on proliferation of stem cells, clonogens, progenitor cells, and resistance remain unknown. Hypothesizing that the overall treatment time available for accelerated proliferation of tumor cells might be a major determinant of outcome for LSCLC, De Ruysscher et al18 performed a systematic overview of studies combining chest irradiation and platinum-based chemotherapy in the primary management of LSCLC and reporting 5-year survival. The overview looked at SER (Start of any treatment until the End of Radiotherapy)-induction, concurrent, sequential radiotherapy and chemotherapy-the study period measured by the year the trial was initiated, and the equivalent radiation dose in 2 Gy fractions, corrected for the overall treatment time of chest radiotherapy (EQD2,T).

Using meta-analysis methodology, the SER was the most important predictor of outcome. There was a significantly higher 5-year survival in the shorter SER arms (overall response [OR]: 0.60; 95% confidence interval [CI] = 0.45-0.80; P = .0006), which was more than 20% when SER was less than 30 days.18 Although no significant relation between the SER and the local tumor control was found (OR: 0.73; 95% CI = 0.46-1.14; P = .16), the local tumor control was higher with increasing EQD2,T radiation doses (P = .02). This suggests that for local tumor control, both time and radiation dose factors are important.

A lower SER was associated with a higher incidence of severe esophagitis (OR:0.47; 95%CI: 0.33-0.66; P < 0001). SER was not statistically associated with pneumonitis (too few events), severe leukopenia, or thrombocytopenia.

The authors suggest that a short SER (less than 30 days) seems associated with improved survival in LSCLC patients. This idea encompasses accelerated proliferation of tumor cells (stem cells? progenitor cells?) during induction chemotherapy and/or protracted radiotherapy schedules, and may point to a more rational design of combined modality treatment in rapidly proliferating tumors, especially small-cell tumors.

Radiotherapy Target Volume

Selection of the target to treat has evolved due to computerized tomography (CT) and treatment planning systems that allow more precise delineation of structures that warrant treatment from anatomical structures where treatment would cause toxicity. Global mediastinal and supraclavicular irradiation dominated thoracic radiotherapy ports through the trials of the 1980s, including the Intergroup trial,10 which used regional nodal irradiation encompassing the entire mediastinum. Relapse patterns rarely report regional nodal failure as first or ultimate site of relapse, and this is especially true of supraclavicular nodal regions. As in other occult positive disease sites, management with systemic chemotherapy applies. Treating uninvolved nodes with radiotherapy may not be necessary and exposing radiographically normal nodal stations to radiation beams adds toxicity risk.

Field size can be a challenge with bulky presentation and may influence sequence of therapy. A cycle or two of chemotherapy may reduce volume sufficiently to allow a reasonable port for concurrent therapy as soon as possible. The target can then be the residual mass after reduction due to the chemotherapy. Data supporting this are limited, but an analysis from the Mayo Clinic19 provides support for the tactic of targeting postchemotherapy residual mass, or at least finds no clear hazard by not encompassing the initial bulk. This issue has not drawn attention sufficient to construct a prospective clinical trial to test issues related to volume.

Targeting and treating only nodal structures that measure 1 cm or larger on CT scan, clinically palpable nodes in the supraclavicular fossa, and disease found by bronchoscopy constitute an appropriate approach. Elective treatment of uninvolved nodes does not have a good rationale and risks exposing normal tissue to consequent toxicity.

Dose, Time, Fractionation

The duration of overlapping chemotherapy and thoracic radiotherapy may influence tolerability and survival as well as toxicity. In the 1980s, the issue was whether thoracic radiotherapy was needed at all, and the responsiveness of small-cell lung cancer to either radiotherapy or chemotherapy was so great that higher doses were not contemplated.  Doses commonly recommended were in the range of 40 to 50 Gy.

Concern for spinal cord tolerance was managed by simple techniques such as a posterior shield. The spinal cord shield also blocked the tumor centrally, but this was considered a necessary problem. Since that time, treatment planning techniques allow delivery of dose to a target defined by radiation oncologists without concern for spinal cord tolerance. Toxicity to the lung and esophagus is a more practical concern. Modern planning allows delivery of daily or twice daily doses higher than ever considered 25 years ago.

Evidence shows that the standard dose and treatment for LSCLC is 45 Gy delivered in 3 weeks in 30 fractions of 1.5 Gy, administered concurrently with cisplatin plus etoposide.10 While the North Central Cancer Treatment Group (NCCTG) reported a study using twice-daily radiotherapy treatments,20 the methodology used varied greatly from the Intergroup trial.5 The trial mandated induction therapy on both arms. The twice-daily radiotherapy treatment incorporated a 2-week interruption in treatment. Even if the reported dose seems higher, the time protraction diminishes the effect and fosters proliferation of resistant clones. Not surprisingly, with both treatment arms having protracted treatment, one delivered less conveniently, the survival was not different.20

In Canada, 40 Gy in 3 weeks is still widely used.21 We really do not know that longer treatments or higher doses are better for local control or survival, but we are now able to deliver doses up to 70 Gy in 7 weeks22 without a clear signal that higher doses are superior. (See Table 3 for a listing of variations in radiotherapy dose, time, and fractionation.)

Click to enlarge

The Massachusetts General Hospital (MGH) group23 has consistently endorsed a policy of higher doses of once daily treatment and slowly escalated the dose in successive cohorts of patients to 70 Gy. Survival plots of patients treated at higher doses show no inferiority and possibly a slight benefit for protracted high dose treatment. 

Both Cancer and Leukemia Group B and MGH data sets use induction chemotherapy and postchemotherapy target volumes. American cooperative groups have escalated LSCLC thoracic radiotherapy doses to 61 to 70 Gy, paralleling doses used for stage III non-small cell lung cancer. None of these doses, however, appears superior to 40 to 45 Gy in 3 weeks, and lengthy high dose treatments over 6 to 7 weeks are associated with a long SER. Additionally, protracted thoracic irradiation overlaps with more chemotherapy cycles, leading to more dose reductions and delays and unnecessary toxicity. There is no evidence from controlled trials that demonstrates benefit from higher dose thoracic irradiation and no such study has been approved.

Prophylactic Cranial Irradiation

About 15% to 20% of SCLC patients present with brain metastasis. This varies with the method of prevalence testing (imaging modality, clinical symptoms, or signs). Patients with cancer control outside the brain have a 50% to 60% actuarial risk of developing brain metastases within 2 to 3 years from diagnosis. Since the brain is wrapped within a “blood brain barrier” impervious to chemotherapy, the brain is a pharmacologic sanctuary, where small-cell cancer cells too few to detect may reside and grow quietly unless eliminated by the spatial cooperation of external radiotherapy.

There really can be no question about the established benefit of prophylactic cranial irradiation (PCI). In a meta-analysis of seven randomized trials evaluating the value of PCI, the risk of developing central nervous system metastases was reduced by >50%.24 Additionally, 3-year overall survival of complete responders (predominately LSCLC) was 20.7% with PCI vs 15.3% in the untreated control group, which is an absolute survival benefit of about 5%.

Where doubt lingers is the prospect of long-term late effects. Everyone has a hoary anecdote of a long-term survivor with a problem, or an MRI scan showing changes. Legends about late toxicity, from patchy alopecia to severe neurocognitive deficits, continue to daunt the routine use of PCI even in complete responders.

The largest of the seven trials comprising this meta-analysis is the 300 patient French trial that randomized complete responding patients to receive 24 Gy in 8 fractions or to observation. Despite the use of 3 Gy fractions, measuring neurocognitive effects through 24 months found a less than 10% frequency of mental testing changes, and that was regardless of whether the patients had received PCI or not.25

Click to enlarge

Recently, the European Organisation for Research and Treatment of Cancer (EORTC) reported a benefit in extensive disease patients who have “responded” to chemotherapy. In this 286 patient prospective randomized trial, the 1-year survival was 23% with PCI and 15% without, with a hazard ratio of 0.76 (CI: 0.59-0.96, P = .02) for the entire period at risk. This trial used 3 and 4 Gy fractions, and demonstrated that the acute effects were transient, but there were too few survivors beyond 24 months to further endorse safety.  All patients entered had received a platinum and etoposide regimen, and were assessed for response after 4 to 6 cycles and then randomized to PCI or observation. Despite 75% having obvious disease remaining in the chest, dealing with the sanctuary site significantly altered survival. (See Table 4.) The EORTC contemplates a trial adding thoracic radiotherapy to see if that too alters survival or pattern of relapse.26

The selection of an optimal dose for PCI that would lead to further decreases in the incidence of brain metastasis with minimal toxicity is the subject of an ongoing international trial. Addressing the question of the optimal PCI dose to prevent metastases, a standard dose of 25 Gy in 10 fractions is being compared to 36 Gy in 18 fractions or 36 Gy in 24 twice daily fractions. Curiously, despite trials using larger than 2 Gy fractions producing survival benefits and no clear acute or late problems, the theory of larger dose per fraction dampens their use, at least in the United States. PCI probably should not be given with systemic chemotherapy because of increased toxicity.27

Conclusion and Future Directions

Paradoxically, the opportunity for demonstrating treatment improvement for SCLC would appear to be greatest for chemotherapy because of the systemic nature of SCLC, but systemic approaches have produced negligible progress. There are no unequivocally better chemotherapy regimens than 4 to 6 cycles of EP. In the LSCLC patient population, the EP regimen is clearly superior to cyclophosphamide plus anthracycline regimens.  The best results for LSCLC occur with initial concurrent EP chemotherapy and accelerated thoracic irradiation.  More intensive protocols or regimens that add another chemotherapeutic agent to the EP backbone have not emerged.  

While chemotherapy advances have been disappointing in SCLC, innovations in radiotherapy from LSCLC papers published during the 1990s demonstrated that a number of radiotherapy interventions had significant survival benefits. These radiotherapy interventions include addition of thoracic irradiation to chemotherapy, early delivery of thoracic irradiation concurrently with chemotherapy, accelerated thoracic irradiation, and prophylactic cranial irradiation for all responders, with larger dose per fraction in extensive disease and encouragement to enter the Radiation Therapy Oncology Group (RTOG) trial for limited disease. Off-study, 25 to 30 Gy in 2 to 2.5 Gy fractions seem prudent.


CME Post-Test and Evaluation