Bryan R. Leigh, MD

ABSTRACT:    A number of randomized clinical trials and    meta-analyses now support the conclusion that combined modality    regimens that include cisplatin (Platinol)-based chemotherapy improve    survival in stage III nonsmall-cell lung cancer (NSCLC) more    effectively than radiotherapy or surgery alone. Depending on the    therapy, chemotherapy may play a cytoreductive role by eradicating    distant micrometastases, a radiosensitizing role by improving local    control, or do both. In general, sequential therapies in which platinum-based    chemotherapy precedes thoracic radiation or surgery have improved    outcome by affecting distant metastases. In contrast, concurrent    chemoradiotherapy utilizing low-dose cisplatin improves survival by    reducing local recurrence without an effect on distant failure rates.    In view of these observations, chemoradiotherapy strategies that    integrate both radiosensitizing agents and dose levels of    chemotherapy that are effective against micrometastases may prove to    be most efficacious. Since distant metastases remain the major site    of failure, it is also likely that more effective chemotherapy or    other systemic antitumor agents are necessary to increase the current    level of response and survival. Fortunately, several new    chemotherapeutic agents are highly effective against NSCLC as well as    potent radiosensitizers. [ONCOLOGY 14(Suppl 5):35-41, 2000]  

   Nonsmall-cell lung cancer (NSCLC) is the most common cause of    cancer-related death in both men and women in the United States each    year. As the recently revised international staging system    indicates,[1] patient survival is directly related to stage. Survival    rates with surgical resection of early stage disease (stage I and II)    have improved. However, this subgroup accounts for only a small    fraction of the 140,000 new cases of NSCLC each year. Even in this    favorable category, 20% to 40% of patients develop recurrence, which    typically manifests as distant metastases. 

   Locally advanced (stage III) disease, which represents approximately    50,000 cases annually, is more common. This category of tumors is    heterogeneous and, for treatment planning purposes, can be divided    into several patient subgroups (Table    1): stage IIIA without mediastinal node involvement (T3 N1),    IIIA N2 (minimal bulk), IIIA N2 (bulky)/IIIB (T4 or N3 without    malignant pleural effusion), and stage IIIB (malignant pleural    effusion). All of these subgroups are now candidates for combined    modality therapy with curative intentexcept in those patients    with malignant pleural effusion (whose natural history and treatment    options mimic those of stage IV [metastatic] disease). 

   A multidisciplinary clinical research effort that integrates    combinations of chemotherapy, radiotherapy, and/or surgery into the    therapeutic approach against advanced NSCLC has evolved within the    past 10 years.[2] In a number of phase III clinical trials, cisplatin    (Platinol)-based chemotherapyeither in combination with    thoracic radiation or administered before surgeryhas improved    survival more effectively than radiotherapy or surgery alone.[3-9] In    addition, a number of new chemotherapeutic agents are now    incorporated into combined modality regimens. These recent advances,    including the rationale and design of ongoing clinical research    studies investigating new treatment regimens, are reviewed on the    following pages.

   A variety of approaches that integrate chemotherapy, radiation,    and/or surgery into combined modality treatment of NSCLC are    illustrated in Table 2. Each    has potential advantages and disadvantages. Both bimodality    (chemotherapy plus radiotherapy, or chemotherapy plus surgery) and    trimodality (chemotherapy plus radiotherapy plus surgery) options are    under investigation. 

   In both options, NSCLC is considered to be a two-compartment model    consisting of a combination of local-regional disease and distant    micrometastases in which thoracic irradiation or surgical resection    addresses the issue of local control, while the use of chemotherapy    (or other systemic antineoplastic agents) addresses occult distant    spread.[2] Additionally, chemotherapy may play a cytoreductive role    locally or a radiosensitizing role, as described above. 

   For the purpose of this discussion, these approaches to combined    modality therapy can be divided into two avenues: (1) definitive    chemoradiotherapy (nonsurgical combined modality therapy), and (2)    preoperative (neoadjuvant or induction) chemotherapy with or without    radiotherapy followed by surgical resection. A third avenue,    postoperative adjuvant therapy, is not detailed in this review.

   Sequential therapy, in which chemotherapy is completed prior to the    administration of radiotherapy, is the most widely recognized    approach to chemoradiotherapy for locally advanced NSCLC. Because    sequential chemoradiation avoids overlapping toxicities, full doses    and optimal schedules of both modalities may be used, but the    duration of therapy is prolonged. In addition, the potential for    chemotherapeutic radiosensitization is lost. By comparison,    concurrent chemoradiation optimizes both the radiosensitizing and    local cytoreductive potential of chemotherapy. There is no delay in    use of either modality, but concurrent therapy entails the risk of    overlapping toxicity. Conceptually, enhanced toxicity with concurrent    therapy may preclude delivery of full doses of either modality,    depending on the chemotherapeutic regimen and radiation dose    schedule. 

   A number of phase III trials have compared sequential chemoradiation    to radiotherapy alone in stage III NSCLC and their overall results    have been subjected to meta-analysis.[3-7,10] In a landmark Cancer    and Leukemia Group B (CALGB) study, a limited 5-week course of    chemotherapy (100 mg/m² cisplatin during weeks 1 and 5 plus    vinblastine (Velban), 5 mg/m²/week during weeks 15)    followed by radiotherapy (60 Gy/30 fractions) was delivered in    clinical stage III patients who had good performance status (01)    and minimal weight loss (< 5%). Results were compared to patients    who received the same radiotherapy alone. Despite the brief duration    of exposure to this platinum-based chemotherapy, survival was    significantly greater in the combined modality treatment arm (

    = .006). Although median survival was modestly improved (13.8 vs 9.7    mo), survival at 2 years (26% vs 13%) doubled.[3] These results were    confirmed in a subsequent Intergroup trial (Radiation Therapy and    Oncology Group [RTOG] 88-08),[4] comparing identical radiotherapy and    sequential chemoradiotherapy arms with those of the previous CALGB    study. 

   A third treatment arm evaluated the role of hyperfractionated    radiotherapy. Median and 2-year survival in the chemoradiotherapy arm    (13.8 mo/31%) were significantly superior to both standard    radiotherapy (11.4 mo/20%) and hyperfractionated radiation (12.3    mo/24%) (log rank, 

 = .03). However, at 5-year follow-up,    survival was less than 10% in all three arms, which substantiates the    need for better chemoradiotherapy strategies.[11]

   A French study in stage III disease by Le Chevalier compared a    control arm of radiotherapy alone (65 Gy) to three monthly cycles of    PCVC (cisplatin, lomustine [CeeNU], vindesine [Eldesine], and    cyclophosphamide [Cytoxan, Neosar]) followed by the same    radiotherapy.[5] At 3 years, 12% of patients treated with sequential    chemoradiotherapy were alive vs 4% of those treated with radiotherapy    alone (

 = .02).[5,6] In this study, the rate of distant    metastasis was significantly lower in the combined modality group (

    < .001). With rigorous restaging that included bronchoscopy, local    control was less than 20% on either treatment arm, which demonstrates    one of the limitations of sequential chemoradiotherapy. 

   Several phase III trials have compared concurrent cisplatin-based    chemoradiotherapy to radiotherapy alone.[12-15] The hypothesis that    improved local control from chemoradiosensitization could improve    survival is supported by a study from the European Organization for    Research and Treatment of Cancer (EORTC), reported by Schaake-Koning    et al.[12] 

   Three treatment arms were evaluated: split-course radiotherapy alone,    split-course radiotherapy administered concurrently with either    weekly cisplatin (30 mg/m²/week), or daily cisplatin (6    mg/m²) on the days of radiotherapy. Both cisplatin-containing    arms showed superior results to radiation alone, with a statistically    significant survival advantage using daily cisplatin over    radiotherapy alone (

   Of particular interest, improved survival with cisplatin was entirely    due to increased control, with a significant difference in 2-year    freedom from local recurrence in those treated in the    radiotherapy-alone arm (19%) vs the chemotherapy treatment arm (30%).    There was no difference in time to development of distant metastasis.    These results suggest that low-dose cisplatin functioned primarily as    a radiosensitizerleading to improved local control but proving    ineffective against occult systemic disease. Conversely, fully    cytotoxic doses of cisplatin-based chemotherapy, sequenced prior to    radiation in the Le Chevalier study, were effective in eradicating    distant micrometastases without an impact on local control.[5,6]

Concurrent vs Sequential Chemoradiotherapy

   Most recently, the results of a Japanese study by Furuse et al    directly compared concurrent vs sequential chemoradiotherapy in stage    III NSCLC.[16] In this randomized study, median and long-term    survival were clearly superior in the concurrent arm (Table    3). Patients on the sequential arm received two cycles of    mitomycin (Mutamycin)/vindesine/cisplatin (MVP) chemotherapy (56 Gy/2    Gy daily) prior to thoracic radiotherapy; patients on the second arm    received split-course thoracic radiotherapy (56 Gy) concurrent with    MVP. Median survival for the concurrent chemoradiotherapy arm was    16.5 months, significantly better than the 13.3 months observed in    the sequential arm (

 = .05). At 3 years, survival was 27%    with concurrent therapy compared to 12.5% with sequential therapy.    Furthermore, toxicity profiles were acceptable in both of these    combined modality approaches, although esophagitis and    myelosuppression were more frequent in the concurrent arm. If these    data are confirmed by a recently completed RTOG trial, concurrent    chemoradiotherapy will become the standard of care for nonsurgical    therapy of stage III disease. Most ongoing clinical trials have    already incorporated concurrent chemoradiotherapy into the study design.

Preoperative Chemotherapy (± Radiotherapy) Followed by    Surgical Resection

   The concept of preoperative or neoadjuvant therapy was initially    explored in pilot studies that demonstrated feasibility, high    response rates, high resectability rates, and acceptable    perioperative morbidity and mortality.[17-23] All studies used a    cisplatin-based chemotherapeutic regimen. In most studies, concurrent    radiation therapy was delivered preoperatively, to a total dose of 30    to 45 Gy.

   Long-term outcome was recently reported for the largest of these    phase II studies (Southwest Oncology Group [SWOG] 8805) and is    particularly instructive.[23,24] In this study, Albain et al treated    126 patients with pathologically confirmed stage IIIA N2 and IIIB    NSCLC with cisplatin (Platinol)/etoposide (VePesid, VP-16) (PE) and    concurrent radiotherapy (45 Gy) before attempting surgical    resection.[24] In addition to a requirement for pathologic    demonstration of N2 status in stage IIIA patients, over one-third of    the patients had pathologically documented stage IIIB disease (T4 or    N3), which makes this study unique. Complete resection following    induction chemoradiotherapy was accomplished in 73% of patients. In    57% of surgical specimens, pathologic complete or near-complete    response (only a few remaining microscopic tumor foci) was achieved,    including 46% of patients whose response by CT scan was judged to be    stable disease.

   The status of N2 disease at the time of surgical resection was the    most significant predictor of subsequent survival. If residual N2    disease was present, median survival was markedly shorter (10 months)    compared with patients in whom N2 disease had been eradicated by    neoadjuvant chemoradiation (30 months). This finding implies that, in    this setting, surgical resection may have been more prognostic than    therapeutic, by identifying patients destined for long-term survival    based on response to preoperative therapy. There was no difference in    long-term survival between stage IIIA (20%) and IIIB (22%) at a    6-year follow-up. Perioperative morbidity and mortality were    acceptable. However, late pulmonary death occurred in approximately    10% of patients following surgery or boost radiotherapy. This trial    provides the basis for the surgical arm of the current Intergroup    study that randomizes patients with stage IIIA N2 disease to either    preoperative chemoradiotherapy (45 Gy) or definitive    chemoradiotherapy (61 Gy).

   Two small phase III trials that reported the superiority of    preoperative chemotherapy followed by surgery compared to surgery    alone have been recently updated in regard to long-term survival (Table    4). [8,9,25,26] In the Roth study, 60 patients with stage III    (N2 or T3) NSCLC were randomized to receive either surgery alone or    induction chemotherapy with three cycles of cyclophosphamide,    etoposide (VePesid, VP-16), and cisplatin followed by surgery.    Selected patients in both groups received postoperative radiotherapy.    Median survival was 11 months for the surgery alone group vs 64    months for the combined modality group (

 = .008). Long-term    follow-up demonstrates survival at 5 years is 36% in the group    receiving preoperative chemotherapy compared to 15% in the group    receiving surgery alone.[25]

   In a similar study design, Rosell and colleagues randomized 60    patients to either surgical resection followed by chest radiotherapy    (50 Gy), or three courses of induction chemotherapy with mitomycin,    ifosfamide (Ifex), and cisplatin (MIC) followed by surgical resection    and radiotherapy.[9] There was a highly significant improvement in    disease-free survival and overall survival in the chemotherapy arm,    which resulted in early closure of the trial. Median survival was 26    months with preoperative chemotherapy vs 8 months with primary    surgery (

 < .001). At 5-year follow-up, 17% remained alive    in the preoperative chemotherapy group, while there were no long-term    survivors with surgery alone.[26] 

   The results of these trials are provocative; however, due to the    small number and select nature of the patients treated, they are not    definitive.[27] In addition, these trials do not address the question    of whether this patient population would fare equally well if treated    with definitive chemoradiotherapy, the subject of the current    Intergroup trial discussed above. Nevertheless, primary surgery    results in poor survival rates in most patients documented    preoperatively with N2 disease and should not be encouraged,    especially in view of the disappointing results of a recent trial of    adjuvant chemotherapy given postoperatively.[28] 

   In this Intergroup study (Eastern Cooperative Oncology Group [ECOG]    3590), patients with surgically resected stage II and III NSCLC were    randomized to receive postoperative cisplatin/etoposide and    radiotherapy or radiotherapy alone. There were no differences in    disease-free survival or overall survival. 

   These data bring into question the use of any type of adjuvant    therapy, chemotherapy or radiation, in a nonprotocol setting in    surgically resected NSCLC. As noted below, it is possible    combinations containing new agents will be more effective in this    regard. Studies randomizing post-resection patients to receive such    regimens compared with observation are underway.

Integration of New Chemotherapeutic Agents

   Although combined modality strategies do result in a positive affect    on survival, a consensus therapy for stage III NSCLC has yet to be    identified. The results of recent studies define new standards of    care in which 20% to 30% of patients with minimal weight loss and    good performance status treated with combined modality therapy    achieve long-term survival. Nevertheless, most stage III patients    still die of recurrent disease, mandating continued evaluation of new    therapeutic approaches.

   Systemic recurrence remains the major cause of treatment failure.    Therefore, new and improved systemic antineoplastic agents (or    improved utilization of currently available agents) remain a high    priority in clinical research studies. Fortunately, there are now    several new chemotherapeutic agents that demonstrate a relatively    high degree of single-agent activity in NSCLCsome with novel    mechanisms of action and resistance, including the taxanes    (paclitaxel [Taxol] and docetaxel [Taxotere]), the vinca-alkaloid    vinorelbine (Navelbine), the topoisomerase-I inhibitor irinotecan    (Camptosar, CPT-11), and the nucleoside analog gemcitabine (Gemzar).[29]

   In metastatic NSCLC, combinations of these new agents with cisplatin    have reportedly improved response or survival rates, or both, when    compared with either cisplatin alone or older cisplatin-containing    regimens.[30] Additionally, all have demonstrated some evidence of    radiosensitizing properties and synergistic antitumor effects in    combination with radiation in preclinical models.[31-34]

   Based on these preclinical data, a series of clinical trials have    been initiated to combine these new chemotherapeutic agents with    radiation. Critical factors considered in the design of these studies (Table    5) include (1) dose of the new agent (low-dose radiosensitizing    or full-dose cytotoxic), (2) schedule of the new agent (daily,    weekly, or less frequently), and (3) sequential or concurrent    administration with radiation, or a combination of both. Because    these new chemotherapeutic agents have most often been combined with    platinum compounds (cisplatin or carboplatin [Paraplatin]) in the    treatment of stage IV disease, new agents and platinum combinations    have also been employed in many chemoradiation trials in stage III disease.

   Paclitaxel is the most extensively investigated combined modality    therapy. In a study by Choy et al, the maximum tolerated dose in a    weekly schedule concurrent with radiotherapy (60 Gy continuous) was    60 mg/m²/week, given by 3-hour infusion.[35] As in most other    studies of new agents in combination with radiation, esophagitis was    dose-limiting. A number of ongoing or recently completed phase II    studies have evaluated paclitaxel in combination with cisplatin or    carboplatin, administered sequentially before radiation or    concurrently with radiotherapy, or a combination of sequential and    concurrent therapy. 

   Docetaxel has been evaluated in the same fashion: in low doses    weekly, concurrently with thoracic radiation, and in a recently    completed SWOG (S 9804) trial, as full-dose consolidation therapy    following concurrent cisplatin/etoposide/radiation.[36] Toxicities    observed in new agent/radiation trials emphasize the need for careful    dose-escalation studies before these regimens are translated into    routine patient management. For example, in Japanese studies    combining irinotecan with concurrent radiotherapy, severe or fatal    radiation pneumonitis has been observed in a small number of    patients.[37,38] Ongoing studies are defining safe dose schedules of    irinotecan given concurrently with radiotherapy.

   Similar findings have been reported in some of the initial trials of    gemcitabine in combination with thoracic radiation.[39] A recently    completed CALGB trial used a randomized phase II design to    investigate three regimens combining new agents and cisplatin    (gemcitabine/cisplatin, paclitaxel/cisplatin, and    vinorelbine/cisplatin) plus radiotherapy. All three regimens were    initially administered as induction chemotherapy for two cycles and    subsequently concurrent with radiotherapy.[40] Toxicity patterns    varied among the three regimens; efficacy was relatively equivalent.

   Although preliminary results are encouraging, toxicity has been    substantial in some studies and, to date, there are no phase III    trials that demonstrate the superiority of any new agent-based    regimen in combination with radiotherapy when compared with standard    cisplatin/radiotherapy regimens. (The study designs for three North    American cooperative group trials in chemoradiation for stage III    disease are depicted in Figure 1.)    An ECOG study addresses the question of improved radiation by    sequencing induction chemotherapy before either standard once per day    or hyperfractionated accelerated radiotherapy (HART). CALGB is    testing the concept of concurrent vs sequential plus concurrent    chemoradiation, where concurrent therapy consists of    radiosensitizing, low-dose chemotherapy. 

   Lastly, an Intergroup trial involving SWOG, North Central Cancer    Treatment Group (NCCTG), and National Cancer Institute of Canada    (NCI-C) will evaluate the novel therapeutic epidermal growth factor    receptor (EGFR) inhibitor ZD1839 (Iressa) compared to placebo    following maximal cytoreduction with chemoradiotherapy. Carefully    designed and conducted clinical trials such as these are needed to    define the role each of these new chemotherapeutic agents (or other    novel classes of antineoplastic agents) will ultimately play in    combined modality approaches to stage III NSCLC.

   These studies clearly illustrate that multimodality therapy improves    therapeutic outcomes in patients with stage III NSCLC compared with    either radiotherapy or surgery alone. Moreover, platinum-based    chemotherapy is the common denominator of all successful approaches    to date. This remains true even though the exact role of surgery and    radiotherapy in local control, the best sequence of modalities, and    optimal chemotherapy and radiotherapy regimens are still not    definedand may well differ for various subsets of stage III    disease. 

   Despite these advances, stage III NSCLC remains a lethal disease in    the majority of patients. The best way to integrate new    chemotherapeutic agents into combined modality therapy remains    unclear. Evidence is also inconclusive as to whether new-agent    combinations are superior to older cisplatin-based regimens. The    studies continue.

   1. Mountain CF: Revisions in the international system for staging    lung cancer. Chest 111(6):1710-1717, 1997. 

   2. Edelman MJ, Gandara DR, Roach M, Benfield J: Multi-modality    therapy in stage III non-small-cell lung cancer. Ann Thorac Surg    61:1564-1572, 1996.

   3. Dillman RO, Herndon J, Seagren SL, et al: Improved survival in    stage III NSCLC: Seven-year follow-up of the CALGB 8433 trial. J Natl    Cancer Inst 88:1210-1215, 1996.

   4. Sause WT, Scott C, Taylor S, et al: RTOG 88-08 and ECOG 4588,    Preliminary results of a phase III trial in regionally advanced,    unresectable nonsmall-cell lung cancer. J Natl Cancer Inst    87:198-205, 1995.

   5. Le Chevalier T, Arriagada R, Quoix E, et al: Radiotherapy alone    versus combined chemotherapy and radiotherapy in nonresectable    nonsmall-cell lung cancer: First analysis of a randomized trial    of 353 patients. J Natl Cancer Inst 83:417-423, 1991. 

   6. Arriagada R, Le Chevalier T, Rekacewicz C, et al: Cisplatin-based    chemotherapy in patients with locally advanced nonsmall-cell    lung cancer: Late analysis of a French randomized trial (abstract).    Proc Am Soc Clin Oncol 16:446a, 1997.

   7. NonSmall-Cell Lung Cancer Collaborative Group. Chemotherapy    in nonsmall-cell lung cancer: A meta-analysis using updated    data on individual patients from 52 randomized clinical trials. BMJ    311:899-909, 1995.

Articles

A number of randomized clinical trials and meta-analyses now support the conclusion that combined modality regimens that include cisplatin (Platinol)-based chemotherapy improve survival in stage III non–small-cell lung

Evolution of Combined Modality Therapy for Stage III Non–Small-Cell Lung Cancer

   The article by Drs. Gordon and Vokes provides the reader with a    comprehensive overview of the treatment of locally advanced    unresectable nonsmall-cell lung cancer (NSCLC) with    chemotherapy and radiation therapy. The authors review pivotal US    trials conducted by the Cancer and Leukemia Group B (CALGB) and the    Radiation Therapy Oncology Group (RTOG) that demonstrate a modest but    reproducible overall survival advantage with induction cisplatin    (Platinol)/vinblastine (Velban) chemotherapy followed by thoracic    radiotherapy, as compared with radiotherapy alone.[1,2] They update    current and recently completed cooperative group trials attempting to    optimize chemoradiotherapy approaches to stage III disease.

   The authors conclude that patients with advanced NSCLC and a    reasonable performance status benefit from potentially curative    combinations of chemotherapy and radiotherapy. Outside of the setting    of a clinical trial, Drs. Gordon and Vokes favor the use of two    cycles of platinum-based induction chemotherapy followed by standard    radiation therapy.

   Although we agree with the authors general conclusions, several    issues are worthy of comment. First, the pivotal CALGB and RTOG    trials that demonstrate a survival advantage for combined-modality    therapy were limited to patients with a reasonable performance status    and minimal weight loss. These eligibility criteria eliminated    approximately 35% of poor-risk patients with locally advanced NSCLC,    who are not likely to tolerate aggressive cisplatin-based    combined-modality therapy. When applying the results of these two    clinical trials to clinical practice, these eligibility criteria are    often overlooked.

   As described by the authors, the Southwest Oncology Group (SWOG)    recently completed a phase II trial of chemoradiotherapy specifically    targeting this poor-risk population.[3] The chemotherapy used in this    trial consisted of two cycles of a modified-dose carboplatin    (Paraplatin)/etoposide regimen administered concurrently with 61 Gy    of thoracic radiotherapy. The median survival of 13 months and 2-year    survival rate of over 20% are encouraging in this compromised patient    subgroup, and were achieved without treatment-related mortality. The    RTOG is now comparing this regimen to radiotherapy alone in a    randomized phase III trial.

   With respect to good-risk patients, which chemoradiotherapy    approaches have the highest likelihood of demonstrating an    improvement in survival time after the diagnosis of locally advanced    NSCLC? We will discuss two categories of clinical trials that address    this question: (1) trials comparing concurrent chemoradiotherapy to    sequential chemoradiotherapy, and (2) trials evaluating altered    fractionation radiotherapy. Finally, we will preview some novel    nonchemotherapeutic anticancer agents that are likely to be    integrated into chemoradiotherapy programs in the future.

   Whether concurrent chemoradiotherapy results in a superior    therapeutic outcome compared to sequential chemotherapy followed by    radiotherapy remains highly controversial. Concurrent chemoradiation    has the potential advantages of enhanced local control and the    elimination of chemotherapy-resistant but radiation-sensitive tumor    clones before the clones metastasize beyond the chest. The major    disadvantage of concurrent therapy is the likelihood of increased    toxicity, which may limit the dose intensity of chemotherapy or radiotherapy.

   Recently, long-term follow-up of a landmark study by the West Japan    Lung Cancer Group (WJLCG) directly comparing concurrent    chemoradiation to sequential chemoradiation was reported.[4] Patients    in both arms of the study received two courses of cisplatin/vindesine    (Eldisine)/mitomycin (Mutamycin) and 56 Gy of thoracic radiotherapy.    The radiotherapy was delivered as a continuous course in the    sequential arm of the trial but as a split-course in the concurrent    arm with a 10-day break after the first 28 Gy.

   Median survival time was 16.4 months for patients receiving    concurrent chemoradiation, significantly better than the median of    13.5 months reported for those receiving chemotherapy followed by    radiotherapy (P < .05). The advantage persisted at 5 years, at    which point the survival rate in the concurrent arm was 16%, as    compared with 9% in the sequential arm. Hematologic and esophageal    toxicities occurred more frequently in the concurrent arm but were manageable.

   As mentioned by Gordon and Vokes, an RTOG trial comparing concurrent    chemoradiation to sequential chemoradiation has recently completed    accrual. Survival results are expected to be available in the year    2000. The RTOG trial differed from the Japanese trial in that a    cisplatin/vinblastine combination was used instead of    cisplatin/vindesine/mitomycin. Also, the thoracic radiotherapy dose    for the RTOG trial was 63 Gy in a continuous course for both    armsa more conventional regimen by US standards.

   The RTOG trial also has a third arm evaluating cisplatin and oral    etoposide administered concurrently with a 1.2-Gy/fraction    twice-daily radiotherapy regimen to a total dose of 69.6 Gy. Because    of the theoretical advantages of concurrent chemotherapy and the    WJLCG results discussed above, nearly all cooperative group trials    for locally advanced NSCLC now employ concurrent chemoradiation.

   Drs. Gordon and Vokes appropriately underscore the potential    advantages of hyperfractionated accelerated radiation therapy. Since    the completion of the first RTOG trial in patients with NSCLC in the    mid-1970s, the standard dose for thoracic radiotherapy has been 60 to    66 Gy in daily 1.8- to 2.0-Gy fractions. This approach has been    reported to provide local control rates as low as 15%, even when    combined with chemotherapy.[5]

   By delivering more frequent radiation treatments, the overall    treatment duration can be reduced. This approach theoretically    reduces tumor cell repopulation during therapy, believed to be a    significant problem in the treatment of NSCLC and other cancers.

   The continuous hyperfractionated accelerated radiation therapy    (CHART) regimen of Saunders et al delivers 54 Gy at 1.5 Gy/fraction    three times daily for 12 consecutive days.[6] When standard chest    radiotherapy to 60 Gy was compared to the CHART regimen in patients    with stage I-III NSCLC, the 2-year survival rate increased from 20%    with standard radiotherapy to 29% with CHART. The incidence of acute    esophageal toxicity was higher with the CHART regimen, but the most    severe esophagitis occurred after therapy was completed and was    manageable with symptomatic medications.

   A more practical, 16-day hyperfractionated accelerated regimen that    provides continuous therapy except on weekends (HART), which delivers    47.6 Gy, appears to as effective as the original CHART regimen and    has been safely employed in a sequential combined-modality program    with carboplatin/

   paclitaxel (Taxol). The Eastern Cooperative Oncology Group (ECOG) is    now conducting a randomized, phase III trial comparing    carboplatin/paclitaxel followed by standard daily thoracic    irradiation to the same chemotherapy followed by HART. As mentioned    above, the RTOG has evaluated twice-daily radiation therapy with    concurrent chemotherapy in a recently completed phase III trial.    These two trials, if positive, may define a new standard of care for    thoracic radiotherapy.

   Because of increased toxicity with hyperfractionated accelerated    radiation regimens, care should be taken to minimize irradiation of    normal tissues, including the esophagus, normal lung, and heart. This    is best accomplished with three-dimensional treatment planning    systems that are currently gaining widespread acceptance in US    radiotherapy centers.

   Despite recent improvements in the treatment of locally advanced    NSCLC, the 5-year survival rate of 10% to 20% seen in selected    patients treated aggressively with chemotherapy and radiotherapy    remains disappointing. Although ongoing clinical trials with newer    chemotherapeutic agents and improved radiotherapy techniques will    likely improve on these results, dramatic increases in survival are    not expected. Instead, incorporation of novel anticancer agents with    unique mechanisms of action may provide the greatest potential for    substantially improved survival. Although beyond the scope of the    Gordon/Vokes article, several of these agents are under study for    NSCLC and deserve mention.

   The matrix metalloproteinase inhibitors have been demonstrated to    reduce tumor cell invasion, angiogenesis, and metastasis. A phase III    intergroup trial (SWOG 9906) is evaluating a twice-daily oral matrix    metalloproteinase inhibitor (Bay-12-9566) as maintenance therapy    after chemoradiation for stage III NSCLC.

   Tirapazamine, a hypoxic cell cytotoxin, will be studied in    combination with chemotherapy and radiotherapy in stage III NSCLC.

   Other trials will evaluate trastuzumab (Herceptin), a monoclonal    antibody directed against a human epidermal growth factor receptor    (HER-2/neu). Trastuzumab is approved for the treatment of    HER-2/neupositive breast cancer. Overexpression of HER-2/neu,    without gene amplification as occurs in breast cancer, is seen in 20%    to 30% of patients with NSCLC.

   Lastly, a monoclonal antibody against the vascular endothelial growth    factor (anti-VEGF) has shown promise in preliminary studies when    combined with chemotherapy.

   In summary, although therapeutic nihilism is no longer justified in    NSCLC, we cannot yet declare victory. We agree with Gordon and Vokes    that standard treatment in the United States today is platinum-based    chemotherapy in combination with thoracic radiotherapy.    Unfortunately, this treatment results in a median survival duration    of approximately 13 months, and a 5-year survival rate of 10% to 20%.

   The good news is that a number of promising new anticancer agents and    treatment techniques are available for testing. Well-designed    clinical trials will need to determine whether and when these new    approaches should be employed in daily practice.

   1. Dillman, RO, Herndon J, Seagren SL, et al: Improved survival in    stage III non-small-cell lung cancer: Seven-year follow-up of Cancer    and Leukemia Group B (CALGB) 8433 trial. J Natl Cancer Inst    88:1210-1215, 1996.

   2. Sause WT, Scott C, Taylor S, et al: Radiation Therapy Oncology    Group (RTOG) 8808 and Eastern Cooperative Oncology Group (ECOG) 4588:    Preliminary results of a phase III trial in regionally advanced,    unresectable non-small-cell lung cancer. J Natl Cancer Inst    87:198-205, 1995.

   3. Lau DH, Crowley JJ, Gandara DR, et al: Southwest Oncology Group    phase II trial of concurrent carboplatin, etoposide, and radiation    for poor-risk stage III non-small-cell lung cancer. J Clin Oncol    16:3078-3081, 1998.

   4. Furuse K, Fukuoka M, Takada Y, et al: A randomized phase III study    of concurrent vs sequential thoracic radiotherapy in combination with    mitomycin, vindesine and cisplatin in unresectable stage III    non-small-cell lung cancer: Preliminary analysis (abstract). Proc Am    Soc Clin Oncol 16:459a, 1997.

   5. LeChevalier T, Arriagada R, Quiox E, et al: Radiotherapy alone vs    combined chemotherapy and radiotherapy in non-resectable    non-small-cell lung cancer: First analysis of a randomized trial in    353 patients. J Natl Cancer Inst 83:417-423, 1991.

   6. Saunders M, Dische S, Barrett A, et al: Continuous    hyperfractionated accelerated radiotherapy (CHART) vs conventional    radiotherapy in non-small-cell lung cancer: A randomised multicenter    trial. Lancet 350:161-165, 1997.

The article by Drs. Gordon and Vokes provides the reader with a comprehensive overview of the treatment of locally advanced unresectable non–small-cell lung cancer (NSCLC) with chemotherapy and radiation therapy. The authors

Bryan R. Leigh, MD

Articles

Evolution of Combined Modality Therapy for Stage III Non–Small-Cell Lung Cancer

Chemoradiation for Locally Advanced, Unresectable NSCLC