Short-Course Induction Chemoradiotherapy With Paclitaxel for Stage III Non-Small-Cell Lung Cancer

Transcription

1 Short-Course Induction Chemoradiotherapy With Paclitaxel for Stage III Non-Small-Cell Lung Cancer Thomas W. Rice, MD, David J. Adelstein, MD, Jay P. Ciezki, MD, Mark E. Becker, MD, Lisa A. Rybicki, MS, Carol F. Farver, MD, Marjorie A. Larto, RN, Eugene H. Blackstone, MD Departments of Thoracic and Cardiovascular Surgery, Hematology and Medical Oncology, Radiation Oncology, Biostatistics and Epidemiology, Anatomic Pathology, The Cleveland Clinic Foundation, Cleveland, Ohio. Background. This study assessed toxicity, tumor response, disease control, and survival after short-course induction chemoradiotherapy and surgical resection in patients with stage III non-small-cell lung carcinoma. Methods. Forty-five patients with stage III non-smallcell lung carcinoma received 12-day induction therapy of a 96-hour continuous infusion of cisplatin (20 mg/m 2 per day), 24-hour infusion of paclitaxel (175 mg/m 2 ), and concurrent accelerated fractionation radiation therapy (1.5 Gy twice daily) to a dose of 30 Gy. Surgical resection was scheduled for 4 weeks later. Postoperatively, a second identical course of chemotherapy and concurrent radiation therapy (30 to 33 Gy) was given. Results. Induction toxicity resulted in hospitalization of 18 (40%) patients for neutropenic fever. No induction deaths occurred. Of 40 (89%) patients who underwent thoracotomy, resection for cure was possible in 32 (71%) patients. Pathologic response was noted in 21 (47%) patients, and 14 (31%) were downstaged to mediastinal node negative (stage 0, I, or II). At a median follow-up of 19 months, 24 patients were alive, 10 with recurrent disease. Of 21 deaths, 16 were from recurrent disease, three were from treatment, and two were unrelated. Recurrent disease was distant in 21 patients, distant and locoregional in 2, and locoregional in 3. The Kaplan- Meier projected 24-month survival is 49%. Projected 24-month survival is 61% for stage IIIA, 17% for stage IIIB (p 0.035); 84% for pathologic responders, 22% for nonresponders (p < 0.001); 83% for downstaged patients (stage 0, I, or II), 33% for those not downstaged (p 0.005); and 63% for resectable patients, 14% for unresectable patients (p 0.007). Conclusions. We conclude that short-course neoadjuvant therapy with paclitaxel (1) has manageable toxicity and a low treatment mortality, (2) results in good tumor response and downstaging, (3) provides excellent locoregional control with most recurrences being distant, and (4) has improved the median survival compared with historical controls. Survival was better in stage IIIA patients, resectable patients, pathologic responders, and patients downstaged to mediastinal node negative disease (stage 0, I, or II). (Ann Thorac Surg 1998;66: ) 1998 by The Society of Thoracic Surgeons Presented at the Poster Session of the Thirty-fourth Annual Meeting of The Society of Thoracic Surgeons, New Orleans, LA, Jan 26 28, Address reprint requests to Dr Rice, The Cleveland Clinic Foundation, 9500 Euclid Avenue, Cleveland, OH ( ricet@cesmtp. ccf.org). Neoadjuvant treatment of stage III non-small-cell lung cancer (NSCLC) using second-generation, cisplatin-based combinations has been studied for over a decade. Recent success in the treatment of stage IV NSCLC with newer agents such as paclitaxel, vinorelbine, and gemcitabine has stimulated many physicians to consider their use in stage III disease. Several of these agents have radiosensitizing potential, which makes them even more attractive in neoadjuvant therapy. Previously we found that a short-course neoadjuvant regimen of cisplatin, etoposide, 5-fluorouracil, and accelerated fractionation radiation followed by surgical resection for poor-prognosis stage III NSCLC was well tolerated [1]. This effective regimen was accomplished in less than 2 weeks and spared patients the protracted treatment duration and toxicities of more complex schedules. Substitution of paclitaxel for the less effective etoposide and 5-fluorouracil represents a logical extension of this work [2]. The purpose of this study was to assess toxicity, tumor response, disease control, and survival in patients with surgically staged, poor-prognosis stage III NSCLC who received a short-course of neoadjuvant paclitaxel-based chemoradiotherapy. Patients and Methods Eligibility Criteria Eligibility for this study required a confirmed diagnosis of NSCLC based on histologic examination of biopsy material or cytologic examination of needle aspiration specimen or bronchoscopic brushings or washings. All patients were examined before treatment by a thoracic surgeon, radiation oncologist, and medical oncologist. Initial staging evaluation included a medical history, physical examination, complete blood cell count, electrocardiogram, urinalysis, and serum chemistry tests that 1998 by The Society of Thoracic Surgeons /98/$19.00 Published by Elsevier Science Inc PII S (98)

2 1910 RICE ET AL Ann Thorac Surg SHORT-COURSE NEOADJUVANT THERAPY FOR NSCLC 1998;66: Fig 1. Treatment schema for short-course neoadjuvant therapy for stage III non-small-cell lung carcinoma. (CDDP cisplatin; RT radiation therapy.) included urea nitrogen, creatinine, calcium, phosphorous, alkaline phosphatase, serum glutamic-oxaloacetic transaminase, lactic dehydrogenase, albumin, total protein, bilirubin, and uric acid. Pretreatment chest x-ray, pulmonary function studies, bronchoscopy, bone scan, and computed tomography (CT) of brain, chest, and abdomen were required to ensure the absence of any hematogenous dissemination. All patients were surgically staged using the 1992 American Joint Committee on Cancer staging criteria for lung cancer [3]. Patients with stage III (M0) carcinoma who had N2 or N3 mediastinal lymph disease or T4 primary tumors (excluding malignant pleural effusion) were eligible. Patients with T3N0 disease were not eligible. Patients were required to meet criteria for medical suitability for pulmonary resection and to have an initial Eastern Cooperative Oncology Group performance status of 0 or 1. Patients who had previous treatment for this malignancy or who had evidence of significant preexisting renal, hepatic, hematologic, or metabolic dysfunction were not entered into the study. Patients with an uncontrolled second primary malignancy were also excluded. This study was approved by the Institutional Review Board of The Cleveland Clinic Foundation (RPC# 4696, initial approval 6/28/94) and was reviewed annually. Written informed consent was required and obtained from all patients before commencement of therapy. Induction Therapy Treatment consisted of a single course of preoperative chemotherapy with a 4-day continuous intravenous infusion of cisplatin (20 mg/m 2 /d) and a 24-hour infusion of paclitaxel, (175 mg/m 2 ) given on day 1 (Fig 1). Chemotherapy was given peripherally and required hospitalization. At least 2 L of intravenous fluid were infused daily, and intensive ondansetron-based antiemetic therapy was administered as required. Concurrent with chemotherapy, accelerated fractionation radiation therapy was given at a dose of 1.5 Gy twice daily, to a total preoperative dose of 30 Gy. Half the total dose was administered during the first week, concurrent with chemotherapy. The second half of the radiation dose was given during the second week. All induction therapy was completed in 12 days. Radiation therapy volumes were dependent on the location of the primary carcinoma. Treatment portals for upper lobe and middle lobe carcinomas included bilateral supraclavicular fossae, mediastinum (to 5 cm below the carina), ipsilateral hilum, and primary carcinoma with a 2-cm margin. Portals for lower lobe lesions included the ipsilateral hilum, the mediastinum (to the diaphragm), and the primary carcinoma with a 2-cm margin. For superior sulcus tumors, the upper border was placed at the C3-C4 interspace and the ipsilateral axilla was also included. Because of expected toxicity after induction therapy, patients were followed up at least weekly. Clinical Response Approximately 4 weeks after completion of induction therapy, clinical restaging was performed with a second CT scan and a second set of pulmonary function studies. Conventional response definitions were used for this CT scan-based clinical response assessment. A complete response was defined as the total disappearance of all radiographic and clinical evidence of disease. A partial response was defined as any response less than complete, but with greater than 50% reduction in the sum of the products of the crossed diameters of all measurable lesions. Patients with less tumor shrinkage were considered nonresponders. Progressive disease was defined as greater than a 25% increase in the sum of the products of the crossed diameters of all measurable lesions or the appearance of new locoregional or metastatic disease. The demonstration of metastatic disease at any time was scored as disease progression, irrespective of any locoregional improvement. All patients who had a clinical response or stable disease proceeded to thoracotomy. Surgical Procedure Mediastinoscopy was not repeated and resection was undertaken regardless of the presence or absence of radiographic evidence of mediastinal lymph node involvement at clinical restaging. A posterolateral musclesparing thoracotomy was used for all patients. Lobectomy was preferred; however, a bilobectomy or pneumonectomy was done for primary tumor invasion or regional lymph node metastases. When necessary, resection of adjacent structures was performed. Unilateral mediastinal lymphadenectomy was accomplished with all pulmonary resections. Pathologic Response Demonstration of a pathologic response postoperatively required pathologic downstaging. Complete disappearance of all tumors at surgery qualified as a complete response. A reduction in either tumor or node status from the pretreatment surgical stage was considered a partial response. Reduction in tumor or node with reciprocal progression in either did not constitute a response. Postoperative Therapy After postoperative recovery, patients were referred for a second course of chemoradiation therapy. A chemotherapy regimen identical to the pretreatment regimen was given concurrent with postoperative accelerated fractionation radiation to a total postoperative dose of 30 to 33 Gy. The total spinal cord dose for both preoperative and postoperative treatment was kept below 45 Gy by using

3 Ann Thorac Surg RICE ET AL 1998;66: SHORT-COURSE NEOADJUVANT THERAPY FOR NSCLC 1911 oblique and lateral portals. No posterior spinal cord blocks were used. Disease Control and Survival After the completion of all treatment, patients were followed up and clinically reevaluated at least every 3 months. Radiographic and endoscopic procedures were repeated as needed. Recurrences were classified as locoregional (inside the ipsilateral thorax), distant (outside the ipsilateral thorax), or both. Survival curves were constructed using the Kaplan-Meier method and compared using the log-rank test. Survival times were calculated from the date of initial chemoradiotherapy. Results were analyzed as of October 31, Results Table 1. Clinical Characteristics and Cancer Staging Characteristic No. % Age (y) Median 59 Range Sex Male Female Race White Black Asian Histology Adenocarcinoma Large cell carcinoma Squamous cell carcinoma 7 16 Stage T1N T2N T3N IIIA T2N T3N T4N T4N IIIB Table 2. Induction Chemoradiotherapy Toxicity Symptom No. % Limits Nausea or vomiting , 94 Severe (grade 3 or 4) , 7 Mucositis or dysphagia , 94 Severe (grade 3 or 4) , 28 Neutropenia 1,000/mm , 90 Hospitalization with fever , 49 Nephrotoxicity , 7 Toxic death 0 0 Between October 1994 and April 1997, 45 patients with stage III NSCLC examined at The Cleveland Clinic Foundation were entered into this clinical trial. Clinical characteristics and staging are given in Table 1. Surgical staging via cervical mediastinoscopy was done in 44 patients. One patient with N3 disease underwent scalene node biopsy only. Thirty-eight patients had T1 (6 patients) and T2 (32 patients) tumors determined by pretreatment chest CT scan. Five patients had T3 tumors defined by CT. Two patients were staged T4: one patient by a combination of CT scan and clinical findings (superior vena caval obstruction) and the other after direct mediastinal invasion was found at mediastinoscopy. Toxicity from induction treatment was significant (Table 2). Although chemotherapy-induced nausea and vomiting was experienced by most patients, it was entirely manageable without treatment interruption. Mucositis and dysphagia were also common, but brief in duration. Neither enteral nor parenteral nutritional support was necessary. Unscheduled hospitalization of 20 patients (44.4%) was required for treatment of toxicity during induction. Eighteen patients were hospitalized for fever and neutropenia. One patient with a normal white blood cell count was admitted with a Staphylococcus epidermidis bacteremia, and one patient had a stroke after induction therapy. No deaths occurred during induction treatment. After induction chemoradiotherapy and before the surgical procedure, a clinical response determination was made on the basis of the second chest CT scan. At this time, 24 patients (53.3%) achieved a partial response. No complete responses were seen. Of the 21 patients (46.7%) who did not demonstrate a clinical response, 15 patients (33.3%) had stable disease and 6 (13.3%) had disease progression. Of 40 operable patients (88.9%), 32 (71.1%) were resectable for cure. Five patients were inoperable: 2 because of distant metastatic disease, 2 because of locoregional disease, and 1 patient became medically inoperable. Eight patients were not resected for cure. Five of these had incomplete resections (five lobectomies) and three had exploratory thoracotomies only. Thirteen pneumonectomies, 1 bilobectomy, and 18 lobectomies accounted for 32 curative resections. One patient required an enbloc chest wall resection with lobectomy. Eight patients (20%) experienced significant postoperative complications (Table 3). Two deaths occurred in the perioperative period; one patient died from respiratory failure secondary to adult respiratory distress syndrome, and the other from sudden death at home. During the operation, a pathologic response was seen in 21 patients (47.7%), with no residual disease found in 5 patients (11.1%) (Table 4). Fourteen patients (31%) were downstaged to mediastinal node negative disease (stage 0, I, or II) (Table 4). A second course of chemoradiotherapy was administered to 36 patients; 3 of 5 inoperable patients, 6 of 8 operable but not curatively resected patients, and 27 of 32

4 1912 RICE ET AL Ann Thorac Surg SHORT-COURSE NEOADJUVANT THERAPY FOR NSCLC 1998;66: Table 3. Postoperative Complications Table 5. Disease Recurrence Complication No. % Limits (%) Recurrence No. % Limits (%) Supraventricular arrhythmia , 17 Pneumonia , 8 Adult respiratory distress syndrome , 8 Bronchopleural fistula , 8 Reoperation for bleeding , 8 None , 44 Distant , 55 Distant and locoregional 2 4 1, 10 Locoregional 3 7 3, 13 curatively resected patients. Nine patients who did not receive postoperative chemoradiotherapy included 2 patients who died perioperatively and 7 patients with distant metastatic disease detected either preoperatively or shortly postoperatively. Toxicity findings during postoperative treatment were similar to those during induction. There was one death during the second course of chemoradiotherapy from a postpneumonectomy bronchopleural fistula and respiratory failure. The planned postoperative radiotherapy dose of 30 to 33 Gy was administered to 34 patients. Two patients developed metastatic disease during their consolidation therapy, which ended their therapy only. At a median follow-up of 19 months, 24 patients (53%) were alive, 10 with recurrent disease. Of 21 (47%) deaths, 16 were from recurrent disease, 3 were from treatment and 2 were unrelated. Disease recurrence patterns are given in Table 5. Locoregional failure occurred in only 5 of the 45 patients (11.1%). Among the five pathologic complete responders, there were no locoregional failures. The projected 2-year Kaplan-Meier survival is 49% (Fig 2) with a median of 22 months. Projected 24-month survival is 61% for stage IIIA and 17% for stage IIIB ( p 0.035). Survival is better for patients who responded to induction chemoradiotherapy, with a 24-month projected survival of 84% for pathologic responders and 22% for nonresponders ( p 0.001) (Fig 3); and 83% for downstaged patients (stage 0, I or II) and 33% for those not downstaged ( p 0.005) (Fig 4). Two-year projected survival is better for patients who had a curative resection than for unresectable patients (63% versus 14%, p 0.007). Comment For patients with multilevel ipsilateral mediastinal lymph node metastases (N2), contralateral or supraclavicular lymph node metastases (N3), or T4 primary tumors, surgical resection is generally unsuccessful and longterm survival is dismal. Similarly, definitive radiation therapy given with curative intent will rarely achieve its goal. Multiple drug chemotherapy treatment alone has a minimal effect on survival. Because of limited success with single modality treatment, multimodality treatment regimens have been suggested for these patients. For these reasons, it is critically important to stage these patients accurately with surgery before making treatment decisions. Although the chest CT scan is valuable to determine the extent of the primary disease, its assessment of the mediastinal lymph nodes might be falsely positive in as many as 30% of cases and falsely negative in 20%. Cervical mediastinoscopy is an excellent mediastinal staging procedure. The presence of positive nodes at the time of mediastinoscopy has been used to define a patient as having unresectable, poor-prognosis stage III disease. All patients in our series had either mediastinoscopy-positive cancers or supraclavicular lymph node-positive biopsies, which insures, with some degree of certainty, the assignment of these patients to the poor prognosis subgroup. The multimodality programs generally used for patients with stage III NSCLC use a number of different chemotherapeutic treatments. Induction chemotherapy Table 4. Pathologic Response No. % Limits (%) Complete response , 18 Partial response , 44 Stable , 40 Disease progression , 28 Not evaluated , 7 Pathologic downstaging , 40 Stage , 18 Stage I , 26 Stage II , 7 Fig 2. The Kaplan-Meier survival curve. The median survival is 22 months with a projected survival of 49% at 2 years. At 1 and 2 years, 70% confidence levels are depicted. The number of patients at risk at 1, 2, and 3 years is recorded in parentheses.

5 Ann Thorac Surg RICE ET AL 1998;66: SHORT-COURSE NEOADJUVANT THERAPY FOR NSCLC 1913 Fig 3. The Kaplan-Meier survival for pathologic responders and nonresponders. At 1 and 2 years, 70% confidence levels are depicted. The number of patients at risk at 1, 2, and 3 years is recorded in parentheses. followed by definitive radiation [4, 5], concurrent chemotherapy and radiation [6], or preoperative chemotherapy without [7, 8] or with radiation therapy [1, 9] have been tested. Altered fractionation chemoradiotherapy schedules have also been tested [10, 11]. Most of these schedules used second generation chemotherapy, generally cisplatin/vinca alkaloid or etoposide based. These aggressive treatment programs resulted in long-term survival for a modest percentage of patients. Although the more aggressive treatment programs that included surgery have resulted in better locoregional disease control, distant disease recurrence remains the single most vexing problem [1, 9]. Most multimodality treatment programs incorporate multiple courses of chemotherapy or extended radiation therapy and result in a protracted treatment duration and considerable toxicity over time. Our experience suggests the potential of achieving similar treatment success using a very short induction that involves a single chemotherapy course and accelerated fractionation radiation [1]. The introduction of new agents into the chemotherapeutic armamentarium will allow the investigation of new multimodality approaches to treat stage III disease. Several of these agents have radiosensitizing properties, which might improve both locoregional treatment success and control of distant metastases. With these goals in mind, paclitaxel was substituted for the etoposide and 5-fluorouracil combination used in our previous work. Our results demonstrate that this treatment combination is well tolerated. Hematologic toxicity is significant but short-lived and easily managed. There were no deaths from toxicity during induction therapy. Eighty-nine percent of this poor-prognosis patient population proved operable and 71% proved resectable for cure. Perioperative morbidity and mortality were not excessive. The pathologic complete response rate was only 11% and might be attributed to the very short time intervals between beginning treatment and surgical resection. Tumor response is better assessed by recognizing that in only 11% of patients did disease recur locoregionally. This result is superior to the locoregional recurrence rate of 90% in the CALGB 8433 trial of induction chemotherapy followed by radiation [4], or to the greater than 50% locoregional recurrence rate experienced by the patients treated with concurrent chemoradiotherapy by Schaake-Koning and associates [6]. Surgical resection was not included in either study. The distant recurrence rate of 51% is disappointing. Even though successful locoregional management might change the pattern of disease recurrence in these patients, the control of distant disease remains suboptimal. Whether additional courses of these or other chemotherapy agents might reduce the incidence of recurrent distant disease remains speculative. Patients with stage IIIA disease who successfully underwent resection for cure or those who achieved a pathologic response did well. Of 14 patients who were pathologically downstaged to mediastinal node-negative disease (stages 0, I, or II), 12 had good results. Only 2 patients died, for a 2-year survival rate of 83%. This figure is not different from the expected survival rate for patients with de novo mediastinal node-negative NSCLC. Methods to identify this subgroup of stage IIIA patients who might benefit from this intervention would be advantageous. Alternative treatments for patients who are unlikely to be treated successfully by these present regimens should be studied. In conclusion, paclitaxel-based concurrent chemotherapy and accelerated fractionation radiation therapy is timesaving, safe, and produces excellent local regional control. Distant disease recurrence remains a significant problem. Further exploration of these newer chemotherapeutic agents in conjunction with radiation is warranted. References Fig 4. The Kaplan-Meier survival for patients downstaged to Stage 0, I, or II and those not downstaged (stage IIIA, IIIB, or IV). At 1 and 2 years, 70% confidence levels are depicted. The number of patients at risk at 1, 2, and 3 years is recorded in parentheses. 1. Rice TW, Adelstein DJ, Koka A, et al. Accelerated induction therapy and resection for poor prognosis stage III nonsmall-cell lung cancer. Ann Thorac Surg 1995;60: Bonomi P, Kim K, Chang A, Johnson D. Phase III trial comparing etoposide cisplatin versus Taxol with cisplatin-g-

6 1914 RICE ET AL Ann Thorac Surg SHORT-COURSE NEOADJUVANT THERAPY FOR NSCLC 1998;66: csf versus Taxol cisplatin in advanced non-small cell lung cancer, an Eastern Cooperative Oncology Group (ECOG) trial. Proc ASCO 1996;15: Beahrs OH, Henson DE, Hutter RVP, Kennedy BJ. Manual for staging cancer, 4th ed. Philadelphia, Pennsylvania: Lippincott, 1992: 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 1996;88: Sause WT, Scott C, Taylor S, et al. Radiation Therapy Oncology Group (RTOG) and Eastern Cooperative Oncology Group (ECOG) 4588: preliminary results of a phase III trial in regionally advanced, unresectable nonsmall cell lung cancer. J Natl Cancer Inst 1995;87: Schaake-Koning C, van den Bogaert W, Dalesio O, et al. Effects of concomitant cisplatin and radiotherapy on inoperable non-small cell lung cancer. N Engl J Med 1992;326: Roth JA, Fosselle F, Komaki R, et al. A randomized trial comparing perioperative chemotherapy and surgery with surgery alone in resectable stage IIIA non-small cell lung cancer. J Natl Cancer Inst 1994;86: Rosell R, Gomez-Codina J, Camps C, et al. A randomized trial comparing preoperative chemotherapy plus surgery with surgery alone in patients with non-small cell lung cancer. N Engl J Med 1994;330: Albain KS, Rusch VW, Crowley JJ, et al. Concurrent cisplatin/etoposide plus chest radiotherapy followed by surgery for stages IIIA (N2) and IIIB non-small cell lung cancer: mature results of Southwest Oncology Group phase II study J Clin Oncol 1995;13: Choi NC, Carey RW, Daly W, et al. Potential impact on survival of improved tumor downstaging and resection rate by preoperative twice-daily radiation and concurrent chemotherapy in stage IIIA non-small cell lung cancer. J Clin Oncol 1997;15: Eberhardt W, Wilke H, Stamatis G, et al. Preoperative chemotherapy followed by concurrent chemoradiation therapy based on hyperfractionated accelerated radiotherapy and definitive surgery in locally advanced non-small-cell lung cancer: mature results of a phase II trial. J Clin Oncol 1998;16: INVITED COMMENTARY These cases are where the action is, so to speak, for locoregionally advanced non small cell lung cancer. Over the years we have, in my opinion, arrived at a consensus that proved stage IIIA non small cell lung cancer should be considered primarily inoperable. That is particularly the case for mediastinoscopy positive patients. To include stage IIIB patients in such neoadjuvant programs is another issue, but there is at least some precedent to include them in certain clinical investigations. The early enthusiasm for preoperative radiotherapy followed by resection was dampened by several welldone randomized clinical trials three decades ago. The advent a decade later of better chemotherapeutic agents for lung cancer and their use in the neoadjuvant setting rekindled our surgical fires, and this interest was whetted by several clinical trials showing a better survival when chemotherapy is combined with radiation. However, early statistical survival projections are not necessarily the equivalent of final results, and we should wait with caution until final results are reported and published before adopting such new treatment protocols as the standard of care. Unfortunately, fewer than 1% of lung cancer surgical patients are involved in randomized clinical trials, and national cooperation is essential to accrue enough patients in a suitable time frame to get meaningful results of treatment for this most common cause of cancer death in North America. There is currently an ongoing intergroup trial (RTOG 9309) comparing neoadjuvant chemoradiation followed by surgical resection with chemoradiation alone, and until that study has enrolled the appropriate number of patients, I think that we should discourage singleinstitution phase II trials for non small cell lung cancer. Willard A. Fry, MD Suite Ridge Ave Evanston, IL by The Society of Thoracic Surgeons /98/$19.00 Published by Elsevier Science Inc PII S (98)