Drs. Belani and Fossella are consultants to and receive grants from Aventis.
Elderly subgroup analysis of a randomized phase III study of docetaxel plus platinum combinations versus vinorelbine plus cisplatin for first-line treatment of advanced nonsmall cell lung carcinoma (TAX 326)
Article first published online: 15 NOV 2005
Copyright © 2005 American Cancer Society
Volume 104, Issue 12, pages 2766–2774, 15 December 2005
How to Cite
Belani, C. P. and Fossella, F. (2005), Elderly subgroup analysis of a randomized phase III study of docetaxel plus platinum combinations versus vinorelbine plus cisplatin for first-line treatment of advanced nonsmall cell lung carcinoma (TAX 326). Cancer, 104: 2766–2774. doi: 10.1002/cncr.21495
- Issue published online: 8 DEC 2005
- Article first published online: 15 NOV 2005
- Manuscript Accepted: 13 JUL 2005
- Manuscript Revised: 11 MAY 2005
- Manuscript Received: 14 DEC 2004
- Aventis Pharmaceuticals, a member of the Sanofi-Aventis Group
- performance status;
Controversy continues over whether elderly patients with advanced nonsmall cell lung carcinoma (NSCLC) should receive platinum-based chemotherapy. TAX 326 reported improved survival with docetaxel–cisplatin (DC) versus vinorelbine–cisplatin (VC) for advanced NSCLC. DC and docetaxel–carboplatin (DCb) were better tolerated than VC. We analyzed the efficacy and toxicity in patients ages < 65 and ≥ 65 years.
Chemotherapy-naive, TNM Stage IIIB–IV NSCLC patients were randomized to DC (docetaxel 75 mg/m2 and cisplatin 75 mg/m2, d1 q3w), DCb (docetaxel 75 mg/m2 and carboplatin area under the concentration–time curve 6 mg/mL.min, d1 q3w), or VC (vinorelbine 25 mg/m2, d1, 8, 15, and 22 and cisplatin 100 mg/m2, d1 q4w).
Of 1218 patients, 401 were age ≥ 65 years (149/118/134 DC/DCb/VC arms). In the elderly, median survival was 12.6 versus 9.9 months, 1-year survival was 52% versus 41%, 2-year survival was 24% versus 17% for DC versus VC, respectively. DCb survival results were similar to those for VC: median, 9.0 months; 1-year, 38%; 2-year, 19%. Survival outcomes were similar between elderly and younger patients across treatment arms. Compared with younger patients, elderly patients reported moderately higher incidences of NCI CTC (version 1.0) Grade 3–4 asthenia, infection, and pulmonary toxicities across treatment arms, and diarrhea and sensory neurotoxicity for cisplatin-containing arms. Most hematologic toxicities occurred with similar incidences between elderly and younger patients, although neutropenia was slightly increased in elderly patients.
First-line docetaxel–cisplatin chemotherapy showed similar activity in elderly and younger patients with advanced/metastatic NSCLC; elderly patients tolerated docetaxel–platinum well despite experiencing slightly more toxicity than younger patients. Cancer 2005. © 2005 American Cancer Society.
Lung carcinoma is predominantly a disease of older adults, with half of all cases observed in patients ages ≥ 65 years.1 Current statistics from the U.S. show that carcinoma of the lung and bronchus is the leading cause of cancer-related death in men and women in the age groups 60–79 and ≥ 80 years.2 The likelihood of developing invasive lung or bronchial cancer is correlated with age: for men, the probability rises sharply from 1 in 92 at age 40–59 years to 1 in 17 at age 60–79 years.2 Nonsmall cell lung carcinoma (NSCLC) accounts for the majority of lung carcinoma cases in older adults. However, clinical studies of chemotherapy in advanced NSCLC cases often recruit patients slightly younger in age than is representative of the disease population. For example, the median age of participants in major Phase III trials has been < 65 years and the trials were not specifically aimed at studying the elderly.3–5 The age discrepancy in the literature is reflected in clinical practice; consequently, the use of aggressive treatments, such as combination chemotherapy, for adults > 65 years has remained controversial. Concern that an older patient may not tolerate aggressive treatment sufficiently to attain clinical benefit can result in age becoming a primary determinant of treatment course in NSCLC.6 The results of a Medicare survey suggest that only 22% of patients age > 65 years in the U.S. receive chemotherapy for metastatic disease.7 Furthermore, a retrospective database analysis of 364 patients at a single institution in Australia showed that 37% of patients ≥ 70 years of age received no active treatment whatsoever, regardless of disease stage, compared with 13% of patients < 70 years of age (P < 0.001).8
Should elderly patients with advanced NSCLC receive chemotherapy? Instrumental variable analysis and propensity scoring in a retrospective patient cohort from the Survival, Epidemiology, and End Results (SEER) tumor registry have indicated that chemotherapy is as effective in patients age > 65 years as it is in younger individuals and that all patients with TNM Stage IV disease—regardless of age—should be considered for such treatment.9 This is further supported by a population-based study involving 4076 patients with NSCLC in The Netherlands that showed that the presence of comorbid conditions (which were prevalent in 73% of 1325 male patients age ≥ 70 years and 61% of 212 female patients age ≥ 70 years) had no independent prognostic effect on survival, whereas disease stage and treatment did.10 To examine the effects of chemotherapy in elderly patients further, we conducted preplanned subgroup analyses of patients age < 65 and ≥ 65 years enrolled in the multinational, randomized Phase III study TAX 326. This study compared two docetaxel–platinum first-line regimens with a first-line vinorelbine–cisplatin regimen (control) in 1218 patients with advanced and/or metastatic NSCLC and a Karnofsky performance status ≥ 70%.11 In the overall population, docetaxel–cisplatin was shown to have a more favorable response rate and survival than vinorelbine–cisplatin, while both docetaxel–cisplatin and docetaxel–carboplatin were better tolerated and provided consistently improved quality of life compared with vinorelbine–cisplatin.
MATERIALS AND METHODS
The study design and methods employed in the TAX 326 study have been presented previously11 and are summarized briefly below.
Main eligibility criteria were: histologically or cytologically confirmed Stage IIIB or IV NSCLC; no previous chemotherapy; Karnofsky performance status ≥ 70%; age ≥ 18 years; adequate bone marrow, renal, and hepatic function; and written informed consent. Main exclusion criteria included: prior treatment with a biologic response modifier or chemotherapeutic agent; previous or concurrent malignant disease; and a history of brain or leptomeningeal metastases.
The study design is shown is Figure 1. Patients were randomized to one of three treatment arms: docetaxel (75 mg/m2) plus cisplatin (75 mg/m2) every 3 weeks; docetaxel (75 mg/m2) plus carboplatin at a dose calculated to produce an area under the concentration-time curve of 6 mg/mL/min administered intravenously (i.v.) every 3 weeks; or vinorelbine (25 mg/m2) plus cisplatin (100 mg/m2) i.v. every 4 weeks. Patients were treated for six treatment cycles or until progressive disease, death, or intolerable toxicity; continued treatment beyond six cycles was permitted at the discretion of the treating physician. Docetaxel-treated patients received oral dexamethasone prophylaxis against fluid retention and hypersensitivity reactions. Antiemetic prophylaxis was given routinely and cisplatin-treated patients received adequate hydration. Secondary granulocyte-colony-stimulating factor prophylaxis against neutropenia was permitted after completion of Cycle 1.
Survival and Toxicity Assessments
Time to death was defined as the time from randomization to the date of death or date of last contact if the patient was lost to follow-up. Toxicity was assessed after every cycle (weekly for hematologic assessments) and graded using National Cancer Institute Common Toxicity Criteria (NCI-CTC), version 1; where NCI-CTC could not be used, toxicities were graded as mild, moderate, severe, or life-threatening. After treatment discontinuation, patients were followed at 2-month intervals until death.
The primary endpoint of the TAX 326 study was overall survival; it was planned that two independent analyses of survival would be performed for each docetaxel combination versus vinorelbine–cisplatin. Survival was analyzed in the intent-to-treat population; safety was analyzed in the treated population.
For the age-specific analyses presented here, patients were stratified into the age groups < 65 years and ≥ 65 years. The age cut-off of ≥ 65 years to define the population of ‘elderly’ patients was arbitrarily chosen based on consensus opinion that it is an acceptable definition of ‘elderly.’ Survival (assessed using the Kaplan–Meier method) and toxicity were analyzed in the vinorelbine–cisplatin arm versus each of the docetaxel-containing arms within the cohort of patients ≥ 65 years.
The results for the overall population of 1218 patients in the TAX 326 study have been published previously.11
Between July 1998 and January 2000, a total of 1218 patients with advanced or metastatic NSCLC from 28 countries and 140 institutions were enrolled and randomized in the TAX 326 study. Of these, 401 patients were ≥ 65 years of age and were distributed among study arms as follows: 149 docetaxel–cisplatin, 118 docetaxel–carboplatin, and 134 vinorelbine–cisplatin. Apart from median age, baseline characteristics were well balanced between the three treatment arms in the overall population (presented previously11) and for the cohorts of patients age < 65 and ≥ 65 years (Table 1).
|Characteristic||Docetaxel 75 mg/m2 + cisplatin 75 mg/m2q3w||Docetaxel 75 mg/m2 + carboplatin AUC 6 q3w||Vinorelbine 25 mg/m2/week + cisplatin 100 mg/m2q4w|
|Age < 65 yrs (n = 259)||Age ≥ 65 yrs (n = 149)||Age < 65 yrs (n = 288)||Age ≥ 65 yrs (n = 118)||Age < 65 yrs (n = 270)||Age ≥ 65 yrs (n = 134)|
|Gender, n (%)|
|Male||182 (70)||112 (75)||202 (70)||90 (76)||204 (76)||98 (73)|
|Female||77 (30)||37 (25)||86 (30)||28 (24)||66 (24)||36 (27)|
|Median age, yrs (range)||56 (30–64)||69 (65–81)||56 (23–64)||69 (65–87)||56 (35–64)||68 (65–80)|
|Karnofsky performance status, n (%)|
|100%||43 (17)||22 (15)||48 (17)||18 (15)||48 (18)||20 (15)|
|80-90%||204 (79)||124 (83)||227 (79)||97 (82)||210 (78)||110 (82)|
|70%||12 (5)||3 (2)||13 (5)||3 (3)||12 (4)||4 (3)|
|Extent of disease, n (%)|
|Stage IIIB||82 (32)||53 (36)||96 (33)||36 (31)||85 (31)||48 (36)|
|Stage IV||177 (68)||96 (64)||192 (67)||82 (69)||185 (69)||86 (64)|
|No. of organs involved|
|1||41 (16)||29 (19)||41 (14)||23 (19)||51 (19)||22 (16)|
|2||128 (49)||71 (48)||141 (49)||58 (49)||111 (41)||69 (51)|
|≥3||90 (35)||49 (33)||106 (37)||37 (31)||108 (40)||43 (32)|
|Prior anticancer therapy, n (%)|
|None||177 (68)||111 (74)||196 (68)||81 (69)||173 (64)||90 (67)|
|Surgery||50 (19)||27 (18)||47 (16)||25 (21)||63 (23)||27 (20)|
|Radiotherapy||17 (7)||7 (5)||20 (7)||7 (6)||16 (6)||6 (4)|
|Surgery + radiotherapy||15 (6)||4 (3)||25 (9)||5 (4)||18 (7)||11 (8)|
In patients ≥ 65 years, the median numbers of cycles administered in the docetaxel–cisplatin, docetaxel–carboplatin, and vinorelbine–cisplatin arms were 5 (range, 1–13), 6 (range, 1–9), and 3 (range, 1–8), respectively; the mean combined relative dose intensities were 0.93, 0.93, and 0.76, respectively. The median number of cycles administered and mean combined relative dose intensities were similar for patients ≥65 years and < 65 years within each treatment group (Table 2).
|Docetaxel 75 mg/m2 + cisplatin 75 mg/m2q3w||Docetaxel 75 mg/m2 + carboplatin AUC 6 q3w||Vinorelbine 25 mg/m2/week + cisplatin 100 mg/m2q4w|
|Age < 65 yrs||Age ≥65 yrs||Age <65 yrs||Age ≥65 yrs||Age <65 yrs||Age ≥65 yrs|
|No. of patients treated||258||148||287||114||268||128|
|Median no. of cycles (range)||6 (1–10)||5 (1–13)||6 (1–10)||6 (1–9)||4 (1–9)||3 (1–8)|
|Median combined relative dose intensity||0.94||0.93||0.93||0.93||0.79||0.76|
|Median cumulative dose, mg/m2||402:379||375:374||380:1814||374:1649||293:378||224:299|
In the cohort of 401 patients age ≥ 65 years, the median survival time in the docetaxel–cisplatin arm was 12.6 months (95% confidence interval [CI]: 10.6–15.4 months) and 9.9 months (95% CI: 8.7–12.2 months) in the vinorelbine–cisplatin arm (Fig. 2); 1-year survival was 52% (range, 45–60%) and 41% (range, 33–49%), respectively, and 2-year survival was 24% (range, 16–32%) and 17% (range, 10–24%), respectively. Survival results for elderly docetaxel–carboplatin-treated patients were similar to those for elderly vinorelbine–cisplatin-treated patients (Table 3).
|Docetaxel 75 mg/m2 + cisplatin 75 mg/m2q3w||Docetaxel 75 mg/m2 + carboplatin AUC 6 q3w||Vinorelbine 25 mg/m2/w + cisplatin 100 mg/m2q4wa|
|Age < 65 yrs (n = 259)||Age ≥ 65 yrs (n = 149)||Age < 65 yrs (n = 288)||Age ≥ 65 yrs (n = 118)||Age < 65 yrs (n = 270)||Age ≥ 65 yrs (n = 134)|
|Median survival, months (95% CI)||11.0 (9.7–12.2)||12.6 (10.6–15.4)||9.7 (8.7–11)||9.0 (7.6–10.3)||10.1 (9.0–11.5)||9.9 (8.7–12.2)|
|One-year survival, % (95% CI)||44 (38–50)||52 (45–60)||37 (32–43)||39 (29–46)||41 (36–47)||41 (33–49)|
|Two-year survival, % (95% CI)||19 (14–25)||24 (16–32)||17 (12–22)||19 (11–27)||13 (9–18)||17 (10–24)|
Within each of the treatment arms, survival outcomes were similar between younger (< 65 years) patients and older patients (Table 3).
Of 401 patients ≥ 65 years, 390 received treatment and were evaluable for safety (148, 114, and 128 in the docetaxel–cisplatin, docetaxel–carboplatin, and vinorelbine–cisplatin arms, respectively). Among this cohort of patients, the safety profiles of the docetaxel treatment groups were shown to be very similar the vinorelbine–cisplatin group.
Table 4 shows Grade 3–4 adverse events by treatment arm and age group; Grade 3–4 adverse events were generally more frequent with vinorelbine–cisplatin than with docetaxel–cisplatin or docetaxel–carboplatin. There was a moderate trend towards increased Grade 3–4 toxicity in elderly versus younger patients across all three treatment arms (Table 4). Specifically, elderly patients experienced a moderately higher incidence of Grade 3–4 asthenia, infection, pain, neurologic, and pulmonary toxicities than younger patients in all three treatment arms. In the cisplatin-containing arms, elderly patients experienced more diarrhea and sensory neurotoxicity than younger patients.
|Docetaxel 75 mg/m2 + cisplatin 75 mg/m2q3w||Docetaxel 75 mg/m2 + carboplatin AUC 6 q3w||Vinorelbine 25 mg/m2/week + cisplatin 100 mg/m2q4w|
|Age < 65 yrs (n = 258)||Age ≥ 65 yrs (n = 148)||Age < 65 yrs (n = 287)||Age ≥ 65 yrs (n = 114)||Age < 65 yrs (n = 268)||Age ≥ 65 yrs (n = 128)|
|Patients with ≥1 adverse event||50.8||53.4||46.3||58.8||56.7||65.6|
Compared with elderly patients on the vinorelbine–cisplatin arm, elderly patients on the docetaxel–cisplatin arm experienced a higher incidence of Grade 1–4 diarrhea (54.1% vs. 23.4%) and peripheral edema (35.8% vs. 18.0%) but less Grade 1–4 vomiting (52.0% vs. 63.3%). Elderly patient receiving docetaxel–carboplatin demonstrated a lower incidence of Grade 1–4 vomiting (34.2%) than vinorelbine–cisplatin, but slightly higher peripheral edema (29.8%) and diarrhea (45.6%). Grade 3–4 hematologic toxicities in patients are shown in Table 5. While a slight increase in the frequency of neutropenia may have occurred in the docetaxel arms, the incidence of Grade 3–4 neutropenia was similar between all three treatment groups for the elderly population. There was a slight increase in febrile neutropenia for the elderly population, occurring in less than 10% of patients in each arm. As with the full safety population,11 more Grade 3–4 anemia occurred in elderly patients treated with vinorelbine–cisplatin.
|Hematologic toxicity n (%)||Docetaxel 75 mg/m2 + cisplatin 75 mg/m2q3w||Docetaxel 75 mg/m2 + carboplatin AUC 6 q3w||Vinorelbine 25 mg/m2/week + cisplatin 100 mg/m2q4w|
|Age < 65 yrs (n = 256)a||Age ≥ 65 yrs (n = 148)||Age < 65 yrs (n = 286)b||Age ≥ 65 yrs (n = 113)c||Age < 65 yrs (n = 267)d||Age ≥ 65 yrs (n = 125)e|
|Leukopenia||98 (38.3)||75 (50.7)||124 (43.5)||73 (64.6)||143 (53.8)||70 (56.0)|
|Neutropenia||181 (70.7)||121 (81.8)||197 (69.6)||97 (86.6)||215 (80.8)||94 (75.2)|
|Febrile neutropenia||8 (3.1)||12 (8.1)||7 (2.4)||8 (7.0)||9 (3.4)||9 (7.0)|
|Thrombocytopenia||8 (3.1)||3 (2.0)||17 (5.9)||11 (9.7)||12 (4.5)||3 (2.4)|
|Anemia||20 (7.8)||8 (5.4)||27 (9.4)||15 (13.3)||62 (23.2)||32 (25.6)|
Fewer elderly patients on the docetaxel–cisplatin (19.5%) and docetaxel–carboplatin (15.3%) arms discontinued treatment owing to an adverse event than on the vinorelbine–cisplatin arm (32.1%). For all treatment groups, the percentage of patients age >65 years discontinuing treatment due to hematologic toxicity mirrored the percentage observed in the full population.
This preplanned subgroup analysis of the cohort of 401 participants ≥ 65 years of age from TAX 326 indicates that elderly patients treated with platinum-based combination chemotherapy have similar survival rates to those of younger (< 65 years) patients from the same study. Notably, elderly patients treated with docetaxel–cisplatin had a median survival of 12.6 months, while this was only 9.9 months for elderly patients treated with vinorelbine–cisplatin. Survival results for elderly patients treated with docetaxel–carboplatin were similar to those receiving vinorelbine–cisplatin. There was a trend towards increased Grade 3–4 toxicity in patients age ≥ 65 years when compared with younger individuals in all treatment arms. Nevertheless, the overall incidence of Grade 3–4 toxicities (in both elderly and younger patients) was lower with docetaxel–cisplatin and docetaxel–carboplatin compared with vinorelbine–cisplatin. Notably, fewer elderly patients in the docetaxel–cisplatin or docetaxel–carboplatin arms discontinued treatment owing to an adverse event than in the vinorelbine–cisplatin arm. It should be noted that our study used a minimum age cut-off of ≥65 years to define the population of ‘elderly’ patients, considered appropriate at the time of the analysis based on epidemiologic literature; however, this is younger than the cut-off of ≥ 70 years that is increasingly being used in oncology trials at present.
The findings of this analysis are congruent with a similar retrospective analysis of patients from the SEER tumor registry that demonstrated similar survival benefits with chemotherapy for elderly patients as compared with younger patients.9 Importantly, prospective studies designed to assess response to chemotherapy in the elderly support the premise that all patients regardless of age can benefit from treatment. However, the issue of the most suitable regimen remains to be clarified. Single-agent therapy with vinorelbine 30 mg/m2, given on Days 1 and 8 for up to six 21-day cycles, increased median survival to 28 weeks from 21 weeks observed with best supportive care (P = 0.03), as reported by the Elderly Lung Cancer Vinorelbine Italian Study (ELVIS) Group.12 Analysis of 161 patients age ≥ 70 years with Stage IIIB or IV NSCLC showed that increased toxicity from chemotherapy was offset by improved quality of life, function, and symptom scores. Several Phase II trials in elderly patients have demonstrated single-agent activity with gemcitabine, with overall response rates of 18–38% and median survival times of 6.8–9 months reported.13–17
The superiority of doublet chemotherapy over a single-agent regimen in patients of all ages with advanced NSCLC was shown in a recent metaanalysis of 65 eligible trials (13,601 patients).18 Doublet therapy significantly improved tumor response (P < 0.0001) and 1-year survival (P < 0.001) compared with single-agent therapy. The metaanalysis did not support the use of a triplet regimen over a doublet regimen.18 However, there is continued controversy over the relative merits of single-agent versus combination therapies in older patients. Randomized Phase III studies from Italy with patients age ≥ 70 years show conflicting survival findings (Table 5). In one of these trials with nearly 700 patients, the vinorelbine–gemcitabine combination failed to show a benefit over either agent administered alone for survival or response rate, and quality of life was similar in all treatment arms.19 Another study20 showed significantly improved survival with gemcitabine–vinorelbine versus vinorelbine alone. The relative risk of death in the combination arm versus single-agent therapy was 0.48 (P < 0.01).20 Considerably fewer patients were enrolled because of early study closure due to apparent clinical advantages (including improved maintenance of quality of life) with combination therapy.
The efficacy and safety of platinum-based combination therapy in elderly patients has yet to be confirmed in prospective studies. Retrospective analysis of elderly patient subsets from randomized studies has drawn criticism owing to the risk of selection bias against older patients in trials where elderly patient recruitment is not specified.21 Perrone et al.21 cautioned that prospective evidence for the safety of cisplatin, in particular, is currently lacking for elderly patients and that prospective Phase II trials investigating modified schedules to reduce toxicity should be performed before evaluating cisplatin-based regimens in Phase III elderly-specific trials. However, recruitment of elderly patients to trials is subject to multiple difficulties, such as an increased incidence of comorbidities preventing patient inclusion, and lack of physician and patient knowledge about the toxicity and benefit of treatment. Indeed, a recent Cancer and Leukemia Group B (CALGB) educational intervention failed in its aim to increase the accrual of older patients to cancer trials, despite using a multifaceted approach targeted at main member institutions of CALGB and its affiliates.22 Therefore, until prospective Phase III data are available, retrospective subgroup analyses may provide useful comparative data on treatment response by patient age,23 provided these data are considered in context.
In a subgroup analysis of a large International Adjuvant Lung Cancer Trial, patients > 64 years appeared to have minimal survival benefit from cisplatin-based adjuvant chemotherapy after complete tumor resection.24 These findings are in contrast to our study and those of others. The results of our study suggest that docetaxel–platinum combinations are active and generally well tolerated in elderly patients. In particular, docetaxel–cisplatin was associated with both improved survival and better tolerability over the control vinorelbine–cisplatin regimen in elderly patients. Compared with younger patients, diarrhea and sensory neurotoxicity were more common in older patients in the cisplatin-based treatment arms. Another subgroup analysis, involving 86 patients age ≥70 years from the Eastern Cooperative Oncology Group (ECOG) 5592 trial of 574 individuals with advanced NSCLC, indicated that there was no difference in the distribution of survival times with cisplatin-based chemotherapy between fit elderly patients and younger (< 70 years of age) patients (P = 0.29), despite a higher incidence of comorbidities in the elderly group.25 Overall tumor response rates (21.5% vs. 23.3%; P = 0.66), median survival (9.05 vs. 8.53 mo), and toxicity in fit elderly patients were similar to those in younger patients,25 as were baseline quality of life and treatment outcome indices (P = 0.12). Leukopenia (P < 0.001) and neuropsychiatric toxicity (P = 0.002) were more common in the older subgroup, but other toxicities were of similar frequency and severity. Furthermore, a comparison of patients age > 75 years with those age 70–75 years in the ECOG 5592 trial showed that, except for a borderline increase in the incidence of leukopenia (P = 0.06), there was no apparent difference in toxicity between these two age groups, although it should be noted that the > 75 years of age group comprised only 24 patients.25 The efficacy of platinum-based doublets in elderly patients has also been investigated in a retrospective analysis of 227 patients age ≥ 70 years from a total of 1139 eligible enrollees in the ECOG 1594 study.26 Despite having significantly more cardiovascular (P = 0.00001) and other noncardiorespiratory (P = 0.008) comorbidities, fit elderly patients fared as well as fit younger patients, with similar overall response rates (24.5% vs. 22.1%; P = 0.76), progression-free survival (3.75 mo vs. 3.71 mo), and overall survival (median survival: 8.25 mo vs. 8.15 mo; 1-year survival: 35.2% vs. 32.8%; P = 0.53).26 However, elderly patients again experienced more toxicity (especially myelosuppression).
The use of weekly, rather than traditional 3-weekly, administration of smaller doses of chemotherapy agents has been investigated as a means of reducing toxicity (in particular, myelosuppression) in elderly patients. Weekly single-agent paclitaxel appears feasible and generally well tolerated in elderly (> 70 years of age) or frail patients considered unsuitable for platinum-based chemotherapy.27 Results from a Phase II study of weekly versus 3-weekly single-agent docetaxel in 96 chemonaive elderly and/or poor performance status patients with advanced NSCLC suggest that hematologic toxicity is lower with the weekly schedule.28 The weekly regimen was better tolerated by octogenarians than the 3-weekly regimen, and the preliminary efficacy results appeared to be comparable between elderly and younger patients.28 Another Phase II study involving 33 chemonaive elderly (≥ 75 years of age) patients treated with docetaxel 20 mg/m2 and cisplatin 25 mg/m2, both administered as weekly infusions, reported acceptable tolerability with no Grade 4 toxicity.29 A comparison of the pharmacokinetics and pharmacodynamics of this weekly schedule of docetaxel–cisplatin in elderly (≥ 75 years of age) and nonelderly (20–74 years of age) patients showed no significant difference in clearance or volume of distribution of docetaxel or unchanged cisplatin between the two age groups, although elderly patients received a lower dose of docetaxel than nonelderly patients (20 mg/m2 vs. 35 mg/m2).30 However, the pharmacodynamic relationship between the docetaxel area under the concentration–time curve (AUC) and the occurrence of neutropenia did differ between the two age groups, with elderly patients being relatively more sensitive to docetaxel exposure. In summary, age-specific analyses have challenged the premise that elderly patients cannot tolerate chemotherapy sufficiently to derive clinical benefit, and have shown that such treatment can be an effective option for patients age ≥ 65 years. Furthermore, studies comparing single-agent versus doublet therapy suggest that chemotherapy doublets should not be withheld in elderly patients solely on the basis of age. Platinum-based therapy provides an important treatment option because it is linked to survival advantages in patients with advanced NSCLC.31 While many clinicians consider platinum-based regimens to be excessively toxic—especially in the elderly—there is increasing evidence from Phase II studies and subanalyses of Phase III studies to suggest that fit elderly patients can achieve similar clinical benefits to younger patients. Our study supports this view and shows that docetaxel–platinum combinations may be active in fit elderly NSCLC patients. Prospective Phase II and Phase III studies conducted specifically in elderly NSCLC patients are needed to confirm the efficacy and safety of cisplatin-based therapy in this patient population and to define the optimum regimen.
- 3Comparison of survival and quality of life in advanced non-small-cell lung cancer patients treated with two dose levels of paclitaxel combined with cisplatin versus etoposide with cisplatin: results of an Eastern Cooperative Oncology Group trial. J Clin Oncol. 2000; 18: 623–631., , , et al.
- 5Single-agent (SA) versus combination chemotherapy (CC) in advanced non-small cell lung cancer (NSCLC): a CALGB randomized trial of efficacy, quality of life (QOL), and cost-effectiveness. [abstract]. Proc Am Soc Clin Oncol. 2002; 21: 1a., , , et al.
- 8Non small cell lung cancer (NSCLC): are elderly patients being under-treated? [abstract]. Proc Am Soc Clin Oncol. 2004; 23: 642., , , et al.
- 16Activity and toxicity of gemcitabine and gemcitabine + vinorelbine in advanced non-small-cell lung cancer elderly patients. Phase II data from the Multicenter Italian Lung Cancer in the Elderly Study (MILES) randomized trial. Lung Cancer. 2001; 31: 277–284., , , et al.
- 26Age-specific subanalysis of ECOG 1594: fit elderly patients (70–80 yrs) with NSCLC do as well as younger patients (<70). [abstract]. Proc Am Soc Clin Oncol. 2003; 22: 639., , , et al.
- 27Weekly paclitaxel (PCT) in non-small cell lung cancer (NSCLC): results in older or frail patients. A Spanish Lung Cancer Group Trial (GECP 00-02). [abstract]. Proc Am Soc Clin Oncol. 2004; 23: 736., , , et al.
- 28A phase II randomized trial of docetaxel weekly or every 3 weeks in elderly and/or poor performance status (PS) patients (pts) with advanced non-small cell lung cancer (NSCLC). [abstract]. Proc Am Soc Clin Oncol. 2004; 23: 627., , , et al.
- 31Non-Small Cell Lung Cancer Collaborative Group. Chemotherapy in non-small cell lung cancer: a meta-analysis using updated data on individual patients from 52 randomised clinical trials. Br Med J. 1995; 311: 899–909.