Study of paclitaxel, etoposide, and cisplatin chemotherapy combined with twice-daily thoracic radiotherapy for patients with limited-stage small-cell lung cancer: a Radiation Therapy Oncology Group 9609 phase II study.

PURPOSE: To determine the response rate, progression-free survival and overall survival, and toxicity of paclitaxel, etoposide, and cisplatin combined with accelerated hyperfractionated thoracic radiotherapy in patients with limited-disease (LD) small-cell lung cancer (SCLC). PATIENTS AND METHODS: LD-SCLC patients with measurable disease, Karnofsky performance score of > or = 70, and adequate organ function who were previously untreated were eligible for the study. Treatment was as follows. In cycle 1 of chemotherapy, concurrent thoracic radiation therapy was administered. In cycles 2 to 4, chemotherapy was administered alone. In cycle 1, chemotherapy consisted of paclitaxel 135 mg/m(2) intravenous over 3 hours on day 1, etoposide 60 mg/m(2) intravenous on day 1 and 80 mg/m(2) orally on days 2 and 3, and cisplatin 60 mg/m(2) intravenous on day 1. In cycles 2 to 4, the paclitaxel dose was increased to 175 mg/m(2), with the etoposide and cisplatin doses remaining the same as in cycle 1. The thoracic radiation therapy consisted of 1.5 Gy in 30 fractions (total dose, 45 Gy) administered 5 days a week for 3 weeks. RESULTS: Fifty-five patients were enrolled onto the study, and 53 were assessable. The major toxicities included grade 3 and 4 acute neutropenia (32% and 43%, respectively) and grade 3 and 4 esophagitis (32% and 4%, respectively). Two patients died as a result of therapy (one died of acute respiratory distress syndrome, and one died of sepsis). There was one late fatal pulmonary toxicity. The median survival time was 24.7 months. The 2-year survival rate was 54.7%. The median progression-free survival time was 13 months, with a 2-year progression-free survival rate of 26.4%. CONCLUSION: Although this therapeutic regimen is effective in the treatment of patients with LD-SCLC, it is unlikely that the three-drug combination with thoracic radiation therapy will improve the survival times compared with the etoposide plus cisplatin chemotherapy regimen with thoracic radiation therapy in LD-SCLC patients.

Quality assurance methods for the first Radiation Therapy Oncology Group permanent prostate implant protocol.

PURPOSE: To report the quality assurance methodology used by the Radiation Therapy Oncology Group in the first cooperative group, multi-institution Phase II trial of transrectal ultrasound guided permanent radioactive implantation of the prostate for definitive treatment of localized adenocarcinoma of the prostate. METHODS AND MATERIALS: Participating institutions were credentialed to participate in this protocol, Radiation Therapy Oncology Group 98-05. International Commission on Radiation Units and Measurements (ICRU) Report 58 was used as the basis for definition of terms. The AAPM's dosimetric prerequisites for low energy interstitial brachytherapy sources were adopted. A nondigital approach to central review was used. The implant dosimetry was recalculated based upon centrally reviewed target volumes by both a radiation oncologist and a diagnostic radiologist. RESULTS: There are differences in the definition of the postimplant prostate between the participating institution, the central review radiation oncologist, and the central review diagnostic radiologist. Thus, there are differences in dose/volume parameters. Six of the 95 patients reviewed did not meet the per protocol criteria based upon information supplied by the participating institution. This increased to 18 cases when using the postimplant target volume defined by the central oncologist and to 23 cases when defined by the radiologist. CONCLUSIONS: This work indicated that there is a need for a central review process of dose-volume analysis within the cooperative group setting. It is indicated that a digital approach to centralized review, which has now been developed, would result in a higher quality and easier review.

Effect of overall treatment time on outcomes after concurrent chemoradiation for locally advanced non-small-cell lung carcinoma: analysis of the Radiation Therapy Oncology Group (RTOG) experience.

PURPOSE: To determine whether overall treatment time affects outcomes after definitive concurrent chemoradiotherapy for locally advanced non-small-cell lung carcinoma (NSCLC). METHODS AND MATERIALS: Data were analyzed from 3 prospective Radiation Therapy Oncology Group trials (RTOG 91-06, 92-04, and 94-10) in which immediate concurrent chemoradiation (cisplatin-based) was the primary therapy for good-performance status Stage III (and selected inoperable Stage II) NSCLC. "Short" overall treatment time (per protocol) was defined as completing treatment within 5 days of plan; other patients were considered to have had "prolonged" treatment time (protocol violation); treatment time was also analyzed as a continuous variable in a multivariate model. Actuarial analysis was performed for overall survival, progression-free survival, freedom from local-regional progression, and toxicity. RESULTS: A total of 474 patients were analyzed. Median follow-up for surviving patients was 6.1 years. Treatment time was delivered per protocol in 387 (82%), whereas 87 patients (18%) had a prolonged treatment time. Long treatment time was significantly associated with severe acute esophagitis. Median survival was slightly better in patients completing treatment on time (19.5 months vs. 14.8 months), but this did not reach statistical significance (p = 0.15) in the univariate analysis. However, in the multivariate analysis of treatment time as a continuous variable, prolonged treatment time was significantly associated with poorer survival (p = 0.02), indicating a 2% increase in the risk of death for each day of prolongation in therapy. Histology (squamous fared worse) and performance status were also significant in the multivariate model. CONCLUSIONS: This retrospective analysis demonstrates a correlation between prolonged overall radiotherapy treatment time and survival in patients with locally advanced NSCLC, even when concurrent chemotherapy is used. Further study of novel radiation-chemotherapy dose/fractionation regimens is warranted.

Phase I study of thoracic radiation dose escalation with concurrent chemotherapy for patients with limited small-cell lung cancer: Report of Radiation Therapy Oncology Group (RTOG) protocol 97-12.

PURPOSE: The purpose of RTOG 97-12 was to determine the maximum tolerated dose (MTD) of thoracic radiation therapy (RT) with concurrent chemotherapy for patients with limited-stage small-cell lung cancer. PATIENTS AND METHODS: Sixty-four patients received four cycles of cisplatin (60 mg/m(2) i.v.) and etoposide (120 mg/m(2) i.v. Days 1-3) (PE), with concurrent thoracic RT starting on Day 1. Thoracic RT was given during the first two cycles with 1.8 Gy/fraction daily to the clinical target volume, followed by thoracic RT to the gross tumor volume b.i.d. for the last 3, 5, 7, 9, or 11 treatment days (total dose 50.4, 54.0, 57.6, 61.2, or 64.8 Gy, respectively). The MTD was based on the dose that produced Grades 3-4 nonhematologic toxicity (mainly esophagitis and pneumonitis) in greater than 50% of patients. RESULTS: After the first 8 patients were enrolled in Arm 1, administration of etoposide was changed from 120 mg/m(2) i.v. on Days 2 and 3 of each cycle to 240 mg/m(2) p.o. for patient convenience as outpatients. Total thoracic RT doses from 50.4 Gy to 61.2 Gy over 5 weeks given with PE were well tolerated. Three of the first 5 patients in the 64.8 Gy arm developed Grade 3 acute esophagitis; the MTD was determined to be 61.2 Gy. Fifty-four (87%) of the 62 evaluable patients achieved a complete (68%) or partial (19%) tumor response. The 18-month survival was 25% for patients receiving 50.4 Gy and 82% for those receiving 61.2 Gy. CONCLUSIONS: The MTD for this accelerated thoracic RT regimen with concurrent PE was 61.2 Gy over 5 weeks.

Comorbidity and Karnofksy performance score are independent prognostic factors in stage III non-small-cell lung cancer: an institutional analysis of patients treated on four RTOG studies. Radiation Therapy Oncology Group.

PURPOSE: To determine the prognostic role of comorbidity in Stage III non-small cell lung cancer (NSCLC) treated definitively with radiotherapy alone. METHODS AND MATERIALS: A total of 112 patients with clinical Stage III NSCLC (American Joint Commission on Cancer 1997) enrolled in four Radiation Therapy Oncology Group studies (83-11, 84-03, 84-07, and 88-08 nonchemotherapy arms) at a single institution were analyzed retrospectively for overall survival (OS) and comorbidity. Of the 112 patients, 105 (94%) completed their assigned radiotherapy. The median assigned dose was 50.4 Gy to the lymphatics (range 45-50.4 Gy) and 70.2 Gy to the primary tumor (range 60-79.2 Gy). Comorbidity was rated retrospectively using the Cumulative Illness Rating Scale for Geriatrics (CIRS-G) and Charlson scales. Karnofsky performance scores (KPSs) and weight loss were prospectively recorded. Because only 8 patients had a KPS of <70, these patients were combined with patients who had a KPS of 70. The OS of this group was compared with that of the patients with better KPSs (>70). RESULTS: The median survival was 10.39 months (range 7.87-12.91). The 2-, 3-, and 5-year OS rate was 20.5%, 12.5%, and 7.1%, respectively. On univariate analysis, clinical stage (IIIA vs. IIIB) was found to be a statistically significant factor influencing OS (p = 0.026), and the histologic features, grade, tumor size as measured on CT scans, age, tobacco use, weight loss >or=5%, and total dose delivered to the primary tumor were not. A KPS of 2 (p <0.0001) were associated with statistically significant inferior OS. Multivariate analysis with clinical stage, KPS, and comorbidity (severity index) of all patients showed that a KPS 2 were independently associated with inferior OS; clinical tumor stage was not found to be an independent prognostic factor. CONCLUSION: KPS and comorbidity are important independent prognostic factors in Stage III NSCLC. Comorbidity should be included in protocols studying advanced stage NSCLC and used for stratification.

Phase III study comparing chemotherapy and radiotherapy with preoperative chemotherapy and surgical resection in patients with non-small-cell lung cancer with spread to mediastinal lymph nodes (N2); final report of RTOG 89-01. Radiation Therapy Oncology Group.

PURPOSE: To compare the outcome of treatment of mediastinoscopy-verified N2 non-small-cell lung cancer treated with induction chemotherapy followed by either surgery or radiotherapy (RT), with both options followed by consolidation chemotherapy. METHODS AND MATERIALS: A randomized Phase III trial for Stage IIIA (T1-T3N2M0) non-small cell lung cancer was conducted by the Radiation Therapy Oncology Group (RTOG) and Eastern Cooperative Oncology Group between April 1990 and April 1994. After documentation of N2 disease by mediastinoscopy or anterior mediastinotomy, patients received induction chemotherapy with cisplatin, vinblastine, and mitomycin-C. Mitomycin-C was later dropped from the induction regimen. Patients were then randomized to surgery or RT (64 Gy in 7 weeks) followed by cisplatin and vinblastine. RESULTS: RTOG 89-01 accrued 75 patients, of whom 73 were eligible and analyzable. Twelve patients received induction chemotherapy but were not randomized to RT or surgery thereafter. Forty-five patients were randomized to postinduction RT or surgery. Of the analyzable patients, 90% had a Karnofsky performance score of 90-100, 18% had weight loss >5%, 37% had squamous cell histologic features, and 54% had bulky N2 disease. The distribution of bulky N2 disease was uniform among the treatment arms. The incidence of Grade 4 toxicity was 56% in patients receiving mitomycin-C and 29% in those who did not. Only 1 patient in each group had acute nonhematologic toxicity greater than Grade 3 (nausea and vomiting). No acute Grade 4 radiation toxicity developed. The incidences of long-term toxicity were equivalent across the arms. Three treatment-related deaths occurred: 2 patients in the surgical arms (one late pulmonary toxicity and one pulmonary embolus), and 1 patient in the radiation arm (radiation pneumonitis). Induction chemotherapy was completed in 78% of the patients. Complete resection was performed in 73% of 26 patients undergoing thoracotomy. Consolidation chemotherapy was completed in 75% of the patients. No statistically significant difference was found among the treatment arms. The overall progression-free survival rate was 53% at 1 year and 17% at 3 years. The median progression-free survival was 14 months. No difference in the 1-year survival rate (70% vs. 66%) or median survival time (19.4 vs. 17.4 months) between the surgery and RT arms. The median survival in the patients receiving induction chemotherapy only was 8.9 months. Mitomycin-C had no impact on survival (p = 0.75). No statistically significant difference was noted in the time to local failure between the surgical and RT arms. CONCLUSION: The patient accrual to this trial made its results inconclusive, but several observations are notable. In this trial, histologic confirmation of N2 disease in the surgical and nonsurgical arms eliminated the usual biases from clinical staging. In this setting, local control and survival were essentially equal between the surgical and RT arms. The 3- and 5-year survival rates of nonsurgical therapy were comparable to published surgical trials of N2 disease.

Comorbidity and KPS are independent prognostic factors in stage I non-small-cell lung cancer.

PURPOSE: To determine the prognostic role of comorbidity in Stage I non-small-cell lung cancer (NSCLC) treated with surgery or radiotherapy (RT). MATERIALS AND METHODS: One hundred sixty-three patients with clinical Stage I NSCLC were analyzed for overall survival (OS) and comorbidity. One hundred thirteen patients underwent surgery (surgical group) and 50 patients received definitive radiotherapy (RT group). Ninety-six percent of the surgical group had lobectomy or pneumonectomy, and negative margins were achieved in 96% of the patients. The median dose to the tumor for the RT group was 61.2 Gy (range 30.8-77.4). The Cumulative Illness Rating Scale for Geriatrics (CIRS-G) and the Charlson scale were used to rate comorbidity. Karnofsky performance scores (KPS) were available in 42 patients; the rest of the scores were determined retrospectively by two physicians independently, with 97% agreement. RESULTS: The OS was 44% for the surgical group and 5% for the RT group at 5 years. Noncancer-related mortality was observed in 31% and 62% of the surgical and RT patients, respectively. On univariate analysis, performed on all patients (n = 163), squamous cell histologic type (p <0.001), clinical Stage T2 (p = 0.062), tumor size >4 cm (p = 0.065), >40 pack-year tobacco use (p <0.001), presence of a CIRS-G score of 4 (extremely severe, CIRS-G4: [+]) (p <0.001), severity index of >2 (p <0.001), Charlson score >2 (p = 0.004), KPS <70 (p <0.001), and treatment with RT (p <0.001) were associated with a statistically significant inferior OS. Multivariate analysis with histologic features, clinical T stage, age, tobacco use, KPS, comorbidity [CIRS-G(4)] and treatment group on all patients showed that squamous cell histology, >40 pack-year tobacco use, KPS <70, and presence of CIRS-G(4) were independently associated with an inferior OS. Treatment modality, T stage, and age did not have any statistically significant effect on OS. Statistically significant differences were found between the surgical and RT groups in Charlson score (p = 0.001), CIRS-G total score (p = 0.004), severity index (p = 0.006), CIRS-G4(+) (p <0.001), KPS (p <0.001), amount of tobacco use (p = 0.002), clinical tumor size (p <0.001), clinical T stage (p = 0.01), forced expiratory volume in 1 s (p = 0.001), and age (p = 0.008), in favor of the surgical group. CONCLUSION: The presence of significant comorbidity and KPS of <70 are both important prognostic factors, but were found to be independent of each other in Stage I NSCLC. Therefore, comorbidity and KPS assessment are recommended when analyzing the prognostic effects of tumor or treatment-related factors on OS.

The effects of comorbidity and age on RTOG study enrollment in Stage III non-small cell lung cancer patients who are eligible for RTOG studies.

PURPOSE: To determine the influence of measured comorbidity in Radiation Therapy Oncology Group (RTOG) combined modality therapy (CMT) study enrollment in Stage III non-small cell lung cancer (NSCLC). METHODS AND MATERIALS: One hundred and seventy-one patients with a Karnofsky Performance Score ≥70 and clinical Stage III NSCLC were analyzed retrospectively for comorbidity, RTOG study eligibility, and enrollment at initial consultation. Effect of comorbidity scores (Cumulative Illness Rating Scale) were tested on patient selection for CMT, RTOG enrollment, and overall survival. RESULTS: Comorbidity (Grade 4; p < 0.005) and use of radiation only (p ≤ 0.001) were associated with inferior survival independent of other factors. Patient selection for CMT was affected by age (≥70, p < 0.001), comorbidity (severity index [SI] > 2, p = 0.001), and weight loss (>5%, p = 0.001). Thirty-three patients (19%) were enrolled in a CMT RTOG study (Group 1). Forty-nine patients (29%) were eligible but not enrolled (Group 2), and 57 (33%) were ineligible (Group 3). The most common ineligibility reasons were weight loss (67%) and comorbidity in the exclusion criteria of the RTOG studies (63%). Group 1 patients were the youngest (p = 0.02), with the lowest comorbidity scores (p < 0.001) and SI (p < 0.001) compared with Groups 2 and 3. Group 3 patients were the oldest with the most unfavorable comorbidity profile. Comorbidity scores (SI >2; p = 0.006) and age (≥70; p = 0.05) were independent factors influencing RTOG study enrollment in patients meeting study eligibility requirements (Groups 1 and 2). CONCLUSIONS: Comorbidity scales could be useful in stratification of patients in advanced lung cancer trials and interpretation of results particularly regarding the elderly population.

Does response to induction chemotherapy predict survival for locally advanced non-small-cell lung cancer? Secondary analysis of RTOG 8804/8808.

PURPOSE: Induction chemotherapy (ICT) improves survival compared with radiotherapy (RT) alone in locally advanced non-small-cell lung cancer (LANSCLC) patients with good prognostic factors. Concurrent chemoradiotherapy (CCRT) is superior to ICT followed by RT. The question arises whether ICT response predicts the outcome of patients subsequently treated with CCRT or RT. METHODS AND MATERIALS: Between 1988 and 1992, 194 LANSCLC patients were treated prospectively with ICT (two cycles of vinblastine and cisplatin) and then CCRT (cisplatin plus 63 Gy for 7 weeks) in the Radiation Therapy Oncology Group 8804 trial (n = 30) or ICT and then RT (60 Gy/6 wk) on Radiation Therapy Oncology Group 8808 trial (n = 164). Of the 194 patients, 183 were evaluable and 141 had undergone a postinduction assessment. The overall survival (OS) of those with complete remission (CR) or partial remission (PR) was compared with that of patients with stable disease (SD) or progressive disease (PD) after ICT. RESULTS: Of the 141 patients, 6, 30, 99, and 6 had CR, PR, SD, and PD, respectively. The log-rank test showed a significant difference (p <0.0001) in OS when the response groups were compared (CR/PR vs. SD/PD). On univariate and multivariate analyses, a trend was seen toward a response to ICT with OS (p = 0.097 and p = 0.06, respectively). A squamous histologic type was associated with worse OS on univariate and multivariate analyses (p = 0.031 and p = 0.018, respectively). SD/PD plus a squamous histologic type had a hazard ratio of 2.25 vs. CR/PR plus a nonsquamous histologic type (p = 0.007) on covariate analysis. CONCLUSION: The response to ICT was associated with a significant survival difference when the response groups were compared. A response to ICT showed a trend toward, but was not predictive of, improved OS in LANSCLC patients. Patients with SD/PD after ICT and a squamous histologic type had the poorest OS. These data suggest that patients with squamous LANSCLC might benefit from immediate RT or CCRT.

MDM2 and Ki-67 predict for distant metastasis and mortality in men treated with radiotherapy and androgen deprivation for prostate cancer: RTOG 92-02.

PURPOSE MDM2 regulates p53, which controls cell cycle arrest and apoptosis. Both proteins, along with Ki-67, which is an established strong determinant of metastasis, have shown promise in predicting the outcome of men treated with radiation therapy (RT) with or without short-term androgen deprivation (STAD). This report compares the utility of abnormal expression of these biomarkers in estimating progression in a cohort of men treated on RTOG 92-02. PATIENTS AND METHODS Adequate tissue for immunohistochemistry was available for p53, Ki-67, and MDM2 analyses in 478 patient cases. The percentage of tumor nuclei staining positive (PSP) was quantified manually or by image analysis, and the per-sample mean intensity score (MIS) was quantified by image analysis. Cox regression models were used to estimate overall mortality (OM), and Fine and Gray's regressions were applied to the end points of distant metastasis (DM) and cause-specific mortality (CSM). Results In multivariate analyses that adjusted for all markers and treatment covariates, MDM2 overexpression was significantly related to DM (P = .02) and OM (P = .003), and Ki-67 overexpression was significantly related to DM (P < .0001), CSM (P = .0007), and OM (P = .01). P53 overexpression was significantly related to OM (P = .02). When considered in combination, the overexpression of both Ki-67 and MDM2 at high levels was associated with significantly increased failure rates for all end points (P < .001 for DM, CSM, and OM). CONCLUSION Combined MDM2 and Ki-67 expression levels were independently related to distant metastasis and mortality and, if validated, could be considered for risk stratification of patients with prostate cancer in clinical trials.

The importance of protein kinase A in prostate cancer: relationship to patient outcome in Radiation Therapy Oncology Group trial 92-02.

PURPOSE: We previously reported that protein kinase A type I (PKA(RIalpha)) overexpression was predictive of outcome in prostate cancer patients treated with radiotherapy (RT) +/- short-term androgen deprivation (STAD) on Radiation Therapy Oncology Group (RTOG) protocol 86-10. Here, we attempt to verify our prior findings and test the hypothesis that the relationship of the length of AD to patient outcome is affected by PKA(RIalpha) overexpression. EXPERIMENTAL DESIGN: There were 313 cases in the RTOG 92-02 study cohort with available tissue and suitable staining by immunohistochemistry. Median follow-up was 10.1 years. The intensity of PKA(RIalpha) staining intensity was quantified manually and by image analysis. Multivariate analyses were done for overall mortality using Cox proportional hazards models and for local failure, biochemical failure, distant metastasis, and cause-specific mortality using Fine and Gray's regression models. RESULTS: The expression levels of PKA(RIalpha), determined by manual and image analysis, were strongly correlated (P < 0.0001). In the multivariate analyses, manual-quantified and image analysis-quantified PKA(RIalpha) staining intensities were independent predictors of distant metastasis (P < 0.01), local failure (P < 0.05), and biochemical failure (P

Microvessel density >or=60 does not predict for outcome after radiation treatment for locally advanced head and neck squamous cell carcinoma: results of a correlative study from the Radiation Therapy Oncology Group (RTOG) 90-03 Trial.

OBJECTIVES: To assess whether microvessel density (MVD), an immunohistochemical marker for tumor vascularity, predicts for radiotherapy (RT) outcome in locally advanced HNSCC patients. METHODS: A total of 459 patients, enrolled on the RTOG 90-03 trial, had biopsy specimens submitted, and a value for MVD determined, prior to definitive RT. 450 patients were analyzable for this study. Tumor microvessels were stained for factor VIII-related antigen using a standard immunoperoxidase method. The mean number of stained microvessel profiles, from three x200 fields containing the highest MVD (hot spot), was recorded as the MVD. A prospective value of >or=60 was chosen as the threshold for high MVD, tumor vascularity. RESULTS: The median follow-up for the analyzable patients with MVD assessment was 22.0 months and 79.1 months for all living patients. There were no differences concerning the pretreatment characteristics between those RTOG 90-03 patients with a value for MVD and those without a value for MVD. Thus, the present study cohort possessed comparable characteristics with the entire RTOG 90-03 population. MVD values ranged from 5 to 80, with a median value of 30. Only 37 of 450 (8.2%) patients possessed an MVD >or=60. There were no outcome differences for patients with MVD <60 versus >or=60 on multivariate analysis for time to local-regional failure (P = 0.89), time to distant metastasis (P = 0.80), disease-free survival (P = 0.46), and overall survival (P = 0.39). CONCLUSIONS: In this large, correlative study, a MVD >or=60, ie, high tumor vascularity, did not predict for outcome in locally advanced head and neck squamous cell carcinoma patients treated with radiotherapy.

Age is independent of comorbidity influencing patient selection for combined modality therapy for treatment of stage III nonsmall cell lung cancer (NSCLC).

OBJECTIVES: To determine the influence of age and comorbidity in patient selection for treatment of stage III NSCLC with combined modality therapy (CMT). METHODS: There were 102 patients with a Karnofsky Performance Score greater than or equal to 70, and clinical stage III NSCLC analyzed retrospectively for comorbidity. All patients received radiotherapy, and 57 (56%) received CMT with sequential and/or concurrent chemotherapy. Comorbidity was rated retrospectively using the Cumulative Illness Rating Scale for Geriatrics (CIRS-G). The effect of an extremely severe comorbidity score on patient selection and overall survival (OS) was evaluated. RESULTS: Presence of a grade 4 comorbidity (P = 0.02) and use of radiation only (P < 0.01) were associated with a statistically significant inferior OS on multivariate analysis, whereas age greater than or equal to 70, clinical stage IIIB, >5% weight loss, and radiation dose >63 Gy were not. Patients receiving CMT were significantly younger (P < 0.001), with less comorbidity (P < 0.001), and weight loss (P = 0.003) compared with patients receiving radiotherapy alone. A multivariate analysis revealed that age (P < 0.001), comorbidity (P = 0.007), and weight loss (P = 0.002) were independent factors influencing patient selection for CMT. CONCLUSIONS: Age effects patient selection for CMT independent of comorbidity and weight loss in patients with stage III NSCLC and good performance status. This might be related to physician's biases regarding tolerability of CMT in the elderly, and might explain under-representation of elderly in clinical trials of lung cancer. Comorbidity assessment should be included in protocols studying locally advanced stage NSCLC and may be useful for stratification.