Fax: (011) 45-35-45-69-66
Performance status rather than age is the key prognostic factor in second-line treatment of elderly patients with epithelial ovarian carcinoma
Article first published online: 28 MAR 2002
Copyright © 2002 American Cancer Society
Volume 94, Issue 7, pages 1961–1967, 1 April 2002
How to Cite
Gronlund, B., Høgdall, C., Hansen, H. H. and Engelholm, S. A. (2002), Performance status rather than age is the key prognostic factor in second-line treatment of elderly patients with epithelial ovarian carcinoma. Cancer, 94: 1961–1967. doi: 10.1002/cncr.10385
- Issue published online: 28 MAR 2002
- Article first published online: 28 MAR 2002
- Manuscript Accepted: 29 NOV 2001
- Manuscript Revised: 26 NOV 2001
- Manuscript Received: 18 SEP 2001
- M. Kristian and Margrethe Kjær Foundation
- Eva and Henry Frænkel Foundation
- Dagmar Marshall Foundation
- epithelial ovarian carcinoma;
- second-line treatment;
- progression free survival;
- overall survival;
Intravenous cytostatic agents as second-line treatment of epithelial ovarian carcinoma have been withheld from many elderly patients because of fear of toxicity. The purpose of the study is to compare the toxicity and efficacy between elderly (older than 65 years of age) and younger (younger than 65 years of age) patients receiving intravenous second-line treatment of epithelial ovarian carcinoma.
This study was a retrospective analysis of 286 consecutive patients with primary epithelial ovarian carcinoma. Inclusion criteria included histopathologically documented International Federation of Gynecology and Obstetrics (FIGO) Stage IC–IV epithelial ovarian carcinoma; first-line treatment with paclitaxel and a platinum analog; intravenous second-line treatment with topotecan 1.0 mg/m2/day for 5 days, every 3 weeks or paclitaxel (175 mg/m2) and carboplatin (AUC 5), every 3 weeks.
One hundred two patients fulfilled the inclusion criteria receiving topotecan (n = 57) or paclitaxel-carboplatin (n = 45) because of refractory or recurrent disease. The patients' age at start of second-line treatment in the younger (n = 68) and the elderly (n = 34) group were median 54.0 years (range, 34.7–64.3) and 69.5 years (range, 65.1–77.2), respectively. In the patient group aged younger than 65 years, initial performance status was more favorable than in patients aged older than 65 years (P = 0.007) whereas the groups were similar in relation to other potential prognostic factors (P > 0.05). For patients aged younger than 65 years, the overall response rate of 50% (95% confidence interval [CI], 37–63%) was similar to the response rate of 44% (95% CI, 27–62%) in patients aged older than 65 years (P = 0.29). The overall survival from the first day of second-line treatment in patients aged younger and older than 65 years were median 13.3+ months (range, 1.2–38.3+) and 11.8+ months (range, 2.0–41.0+), respectively (P = 0.25). In a multivariate Cox analysis, performance status at time of first-line treatment (0 vs.1–2; P = 0.013; hazard ratio [HR], 2.12), performance status at time of second-line treatment (0 vs. 1–2; P = 0.004; HR, 2.47), and response to second-line treatment (CR + PR vs. NC + PD; P < 0.001; HR, 4.38) were found to be independent significant factors for overall survival whereas age (younger than 65 years vs. older than 65 years) yielded no independent information (P = 0.90). No differences in the rate of postponement of treatment, neutropenia Grade 4, trombocytopenia Grade 3–4, nor hypersensitivity reaction to either cytostatic agent between older and younger patients were noticed (P > 0.05).
Modality of second-line treatment of epithelial ovarian carcinoma should be determined more by assessment of performance status than age per se. Second-line treatment with topotecan or paclitaxel-carboplatin can be safely administered in the aged. Cancer 2002;94:1961–7. © 2002 American Cancer Society.
Thirteen percent of the world population is currently older than 65 years of age, and that figure is estimated to increase to 20% within the next 3 decades.1, 2 Therefore, the problem of treating elderly patients with malignant disease will become more and more relevant. Epithelial ovarian carcinoma, a disease of the elderly, is a chemosensitive disease with response rates of 70–80% to first-line chemotherapy with paclitaxel and a platinum compound after initial cytoreduction.3 Most patients have refractory disease or develop recurrent disease, and second-line therapy should be considered for salvage treatment.3, 4 As second-line therapy, many newer intravenous cytostatic agents such as topotecan, gemcitabine, liposomal doxorubicin, vinorelbine, and docetaxel have shown activity with response rates ranging from 18% to 32%.4 Generally, patients with refractory or recurrent disease are considered incurable, and therefore the benefit of treatment must be balanced toward the potential toxicity of the treatment.5
Many aged patients may have been offered older oral alkylating agents rather than intravenous second-line treatment because of fear of toxicity.4, 6 To elucidate this topic, we found it important to examine the toxicity and efficacy of modern intravenous second-line treatment in epithelial ovarian carcinoma as related to age.
MATERIALS AND METHODS
At the Finsen Center since 1994, combination chemotherapy with paclitaxel and a platinum analog has been the first-line treatment of epithelial ovarian carcinoma International Federation of Gynecology and Obstetrics (FIGO) Stage IC–IV as standard treatment or in Phase II and III studies.7 With respect to second-line treatment, it has been standardized as follows. Patients with progression free interval longer than 6 months after end of first-line treatment have been considered platinum-sensitive, and they were offered reinduction with paclitaxel-carboplatin.8 Patients with progression free interval less than 6 months after end of first-line treatment were classified as platinum-resistant, and since 1997 the standard treatment for these patients has been intravenous topotecan. Patients with platinum-resistant disease included 1) patients with progression of disease during first-line treatment, 2) patients with persistent disease after end of first-line therapy, and 3) patients who responded and subsequently experienced recurrence within 6 months after discontinuation of first-line therapy.
Eligibility criteria included the following: 1) histopathologically documented FIGO IC–IV epithelial ovarian carcinoma, 2) first-line treatment with paclitaxel and a platinum compound, 3) refractory or recurrent disease diagnosed by ultrasonography, computed tomography scans, biochemical parameters, histology, or a combination of these methods (For patients with elevated CA 125 and no morphologic evidence of disease, a confirmed increase of serum CA 125 level to more than twice the upper limit of normal (< 35 U/mL) was used.),9 and 4) intravenous second-line treatment with paclitaxel-carboplatin or single topotecan.
Standard primary therapy consisted of paclitaxel 175 mg/m2 as a 3-hour infusion followed by a 30-minute infusion of carboplatin at an area under the concentration (AUC) time curve of 5 repeated every 3 weeks. Standard recurrence treatment of platinum-sensitive patients was identical to the standard primary therapy. Standard dose of topotecan in patients with platinum-resistant disease was 1.0 mg/m2/day as an 30-minute infusion given daily for 5 consecutive days repeated every 3 weeks. A reduced starting dose of 1.0 mg/m2/day was chosen in preference to the approved dose of 1.5 mg/m2/day because some patients' deaths were attributed to topotecan-induced sepsis in previous studies using the latter dose.10, 11 Dose alterations in both treatments were performed according to hematologic nadir and nadir duration as follows. Paclitaxel and carboplatin dose levels were decreased to 135 mg/m2 and AUC 4, respectively, in the case of trombocytopenia Grade 3–4 or prolonged neutropenia Grade 4 over 1 week using standard World Health Organization (WHO) toxicity criteria.12 If febrile neutropenia or severe bleeding episodes occurred, doses were decreased to 110 mg/m2 and AUC 3, respectively. Dose escalations of paclitaxel and carboplatin were not performed. The dose level of topotecan was decreased to 0.75 mg/m2/day in the case of trombocytopenia Grade 3–4 or neutropenia Grade 4. Dose escalation of topotecan to 1.5 mg/m2/day was performed if myelotoxicity Grade 2 or less has occurred in the recent treatment cycle. Dose reductions in both treatments were also prescribed with nonhematologic toxicity Grade 3–4. Hematologic normalization, i.e., neutrophil count greater than 1.5 109/L (paclitaxel-carboplatin) or greater than 1.0 109/L (topotecan) and platelet count greater than 100 109/L, was required at the onset of a following cycle; otherwise, the treatment was postponed 1 week.
Duration of treatment was dependent on response. All patients were continued on therapy unless progression occurred. Chemotherapy generally was continued to 2 cycles after complete response (CR) has been achieved to a maximum of 12 cycles. Thereafter, the patients were observed with monthly CA 125 values and trimonthly clinical examination after last treatment.
Standard WHO response criteria with modifications were used to verify response.12 A CR was defined as complete resolution of all tumor appearance and normalization of serum CA 125 for at least 1 month. A partial response (PR) was defined as a decrease of at least 50% in the product of the dimensions of all measurable lesions. No change (NC) was defined as bidimensionally measurable lesions decrease of less than 50% or increase of less than 25% in existing lesions. Progressive disease (PD) was defined as an increase of at least 25% in the existing lesions or the identification of new lesions. In patients with measurable disease, response was assessed by ultrasonography and clinical examination after every two courses of chemotherapy. In patients without measurable disease and elevated serum CA 125, the response criteria reported by Rustin et al. were applied.13 Briefly, a response has occurred if there is a 50% decrease in serum CA 125 samples.
Time to progression from start of first-line treatment was defined as the time interval from start of first-line combination chemotherapy to the first objective measurement of progression. Progression free interval was defined as the time interval from end of first-line combination chemotherapy to the first objective measurement of progression. Primary end point for efficacy of second-line treatment was overall survival defined as the time interval between the first day of intravenous second-line treatment to death or the date of analysis (April 1, 2001).
The differences in patient characteristics, response rates, and rates of toxicity in different patient subgroups were compared using the Fisher exact test. Univariate Kaplan–Meier estimates of overall survival were generated for the total patient group for the following potential prognostic factors: FIGO stage; histology; tumor grade; residual disease after staging operation; initial performance status; response to first-line treatment; progression free interval; number of disease sites, tumor size, and performance status at start of second-line treatment; response to second-line treatment; and age at start of second-line treatment. The log-rank test was used to identify significant variables. The significant factors from the univariate analysis were included in a multivariate Cox regression model to find independent prognostic factors for overall survival. The SSPS statistical package (version 10.0; SAS, Chicago, IL) was used. P value less than 0.05 was considered statistical significant.
In the period of August 1994 to June 2000, 286 patients with epithelial ovarian carcinoma were consecutively treated with paclitaxel and a platinum analog after initial staging operation. Of these, 102 patients fulfilled the inclusion criteria having received intravenous second-line treatment with topotecan (n = 57) or paclitaxel-carboplatin (n = 45) because of refractory or recurrent disease. The patients were analyzed by age in two groups. In Group A, patients were younger than 65 years old (n = 68), and in Group B patients were 65 years old or older (n = 34) at time of start on second-line treatment. Patients characteristics are summarized in Table 1. In the patient group aged younger than 65 years, initial performance status was more favorable than in patients aged older than 65 years (P = 0.007) whereas performance status at start of second-line therapy were similar (P > 0.05). The groups were similar in relation to other potential prognostic factors (P > 0.05).
|< 65 yrs||> 65 yrs||P value|
|Patients included (total)||68||34|
|Tumor grade (low)||29||13||NS|
|Residual disease after staging operation (> 1 cm)||49||25||NS|
|Performance status at first-line treatment|
|Tumor size at second-line treatment (cm)|
|Performance status at second-line treatment|
The 102 patients received a median of 6 cycles (range, 2–14) of combination chemotherapy as first-line treatment. Median time to progression from start of first-line treatment was 13.6 months (range, 1.5–45.0). The progression free interval from end of first-line treatment for patients younger and older than 65 years of age were median 6.0+ (range, 0.4–41.7+) and 8.1 (range, 0.5–26.5) months, respectively (P = 0.87). The patients age at start of second-line treatment in the younger and the elderly group were median 54.0 years (range, 34.7–64.3) and 69.5 years (range, 65.1–77.2), respectively. Median number of second-line cycles was 6 (range, 2–16) for patients aged younger than 65 years and 6 (range, 2–12) for patients aged older than 65 years. Treatment results of second-line therapy in relation to age are depicted in Table 2. Two patients were not evaluated for response because of nonmeasurable and nonevaluable disease after recurrence surgery, leaving 100 patients to be evaluated for objective response. For patients aged younger than 65 years, the overall response rate of 50% (95% confidence interval [CI], 37–63%) was similar to the response rate of 44% (95% CI, 27–62%) in patients aged older than 65 years (P > 0.05). The overall survival from first day of second-line treatment in patients aged younger and older than 65 years were median 13.3+ months (range, 1.2–38.3+) and 11.8+ months (range, 2.0–41.0+), respectively (P = 0.25).
|Response||< 65 yrs||> 65 yrs|
|CR + PR||33||50||15||44||NS|
In a univariate analysis of all 102 patients, the following significant prognostic factors for overall survival were identified (P < 0.05): residual disease greater than 1 cm, residual disease greater than 5 cm, initial performance status less than 1, response to first-line treatment, progression free interval longer than 6 months, number of disease sites greater than 1, performance status less than 1 at time of second-line therapy, and response to second-line treatment. No prognostic significance to overall survival was found for advanced FIGO stage (I+II vs. III+IV; P = 0.54), serous histology (P = 0.32), low tumor grade (P = 0.78), tumor size greater than 1 cm at start of second-line treatment (P = 0.19), and age older than 65 years at time of second-line treatment (P = 0.25). As independent prognostic factors for overall survival from start of second-line treatment, the following 3 factors were identified in the Cox analysis: performance status at time of first-line treatment (0 vs.1–2; P = 0.013; hazard ratio [HR], 2.12; 95% CI, 1.17–3.83), performance status at time of second-line treatment (0 vs.1–2; P = 0.004; HR, 2.47; 95% CI, 1.32–4.62), and response to second-line treatment (CR + PR vs. NC + PD; P = 0.0001; HR, 4.38; 95% CI, 2.17–8.81). Figure 1 shows the overall survival curves regarding performance status at start of second-line treatment. The median survival time in patients with performance status 0 (n = 57) and performance status 1–2 (n = 44) at start of second-line treatment were 8.5 and 33 months, respectively (P < 0.0001). At the time of analysis, April 1, 2001, 11 patients (11%) were alive with no evidence of disease, 30 (29%) were alive with disease, and 61 (60%) had died with disease. No patients had died of intercurrent diseases.
Table 3 presents toxicity data. Failure to achieve hematologic recovery within 21 days resulting in a subsequent delay of 1 week in treatment was noted in 26% (95% CI, 15.5–39.7%) (topotecan) and 27% (95% CI, 14.6–41.9%) (paclitaxel-carboplatin) of patients. In patients receiving topotecan, no overall difference in the rate of postponement of treatment, neutrocytopenia Grade 4, nor trombocytopenia Grade 3–4, between older and younger patients was noticed (P > 0.05). In patients receiving paclitaxel-carboplatin, there was no overall difference in the rate of postponement of treatment, nor hypersensitivity reactions to paclitaxel or carboplatin between older and younger patients (P > 0.05). Nonmyeloid toxicity was generally mild.
|Patients||< 65 yrs||> 65 yrs|
|Neutropenia Grade 4||6||17||8||36||NS|
|Trombocytopenia Grade 3–4||8||23||3||14||NS|
|Neutropenia Grade 4a||3||9||0||0||—b|
It is well known that ovarian carcinoma mortality is decreasing in younger women whereas the mortality in women older than 65 years is steadily increasing.2, 14 An epidemiologic study has suggested that the improved survival from noncancer causes and its effect on the surviving gene pool in the aged may account for some of the increase in ovarian carcinoma mortality in the elderly.15 Difference in tumor biology and impaired immunologic response in the aged are among other intrinsic factors suspected to be responsible for the difference in outcomes in various age groups.6, 16 These reasons together with the concept that chemotherapy in general is tolerated poorly in elderly patients have contributed to a more conservative approach in the management of epithelial ovarian carcinoma in the aged.17, 18 However, data from the SEER and other epidemiologic studies have demonstrated that elderly patients with ovarian carcinoma are more likely to have advanced disease diagnosed, that these patients received less surgical and cytostatic primary treatment, and that the overall conservative treatment approach contributed to the decreased survival of elderly ovarian carcinoma patients.19, 20 Hence, currently the general opinion is that elderly shall be offered active primary treatment of epithelial ovarian carcinoma and that discrimination should not be based on age factor alone.4, 6, 21
This study demonstrates that this treatment rationale is also applicable in second-line treatment of epithelial ovarian carcinoma showing no differences in overall survival or overall response rates nor number of toxicity events between comparable groups of older (older than 65 years) and younger (younger than 65 years) patients (P > 0.05). Most studies on the impact of age in the treatment of epithelial ovarian carcinoma focus on primary treatment demonstrating that older patients tolerate radical surgery and cytostatic first-line therapy as well as their younger counterparts.22–24 Only one study focuses on recurrent disease, finding no difference in toxicity and administered dose intensity of paclitaxel (250 mg/m2, every 3 weeks) between patients aged younger and older than 60 years. Progression free and overall survival were not reported in this study.25 The elderly thus should benefit from the full extent of recent advances in second-line treatment of epithelial ovarian carcinoma, and the question is how to select patients for optimal second-line therapy.
The identification of prognostic parameters may lead to grouping of patients that will benefit from targeted therapy.26 In second-line treatment of epithelial ovarian carcinoma, histology, tumor grade, residual tumor, number of disease sites, initial performance status, and time to first recurrence or progression all have been found to be independent prognostic factors for outcome.26, 27 In this study, performance status at time of first-line treatment and performance status at time of second-line treatment were found to be independent prognostic factors for overall survival. This indicates that in second-line therapy the choice of modality of treatment should be based on an individual estimation of the patients physical condition with special focus on performance status.
Differences in prognostic factors between patient groups may affect comparison of outcome. In the patient group of age younger than 65 years, initial performance status was significantly more favorable than in patients aged older than 65 years (P = 0.007) whereas the groups were similar in relation to other potential prognostic factors (P > 0.05; Table 1). This finding may not explain the similar outcome regarding response and survival time in the age groups because low performance status is positive related to outcome, thus making a potential difference even larger. In this study, the patient groups were not corrected for age-adjusted mortality because of the general good performance status of the aged, and none of the patients in the follow-up period died of intercurrent disease.
Glomerular filtration rate declines progressively with age, and aging may lead to changes in other pharmacokinetic parameters that may lead to enhanced drug toxicity.28 Furthermore, second-line treatment has been suspected of causing cumulated toxicity especially in the aged. In this analysis, the frequency of toxicity events were modest. Nadir fever and hypersensitivity reactions to chemoagents were registered in overall 3.9% (95% CI, 1.1–9.7%; 4 of 102) and 11.8% (95% CI, 6.2–19.7%; 12 of 102) of patients, respectively (Table 3). No fatal reactions were observed. No difference in the rate of myelotoxicity or hypersensitivity reactions between elderly and their younger counterparts were noticed. There are no randomized studies comparing toxicity events between newer intravenous chemotherapeutic agents with older alkylating agents, thus making direct comparison difficult because of different inclusion criteria and definitions used. A recent Phase II study on oral treosulfan 500 mg/day in second-line treatment found no incidents of nadir fever, but response rate was as low as 3%.29 The determination of treatment modality in patients with refractory or recurrent disease therefore must be determined by realistic information concerning efficacy, side effects, and schedule. Furthermore, factors such as comorbid conditions, prior toxicity, and patient preferences should be considered. The concept of ovarian carcinoma recurrence treatment as salvage therapy in which treatment options focuses on palliation of symptoms and enhancement of quality of life must be emphasized. Despite modest outcomes and survival as verified in the current study, most patients with recurrent ovarian carcinoma wish to receive aggressive treatment of their cancer.30 However, postponement of treatment because of prolonged nadir was reported in overall 26.5% (95% CI, 18.2–36.1%; 27 of 102) of patients (Table 3), and the approved 5-day schedule of topotecan may be inconvenient to many elderly patients. Future studies on second-line treatment focusing on age thus should include quality-of-life assessment in comparing various treatment options as recommended by others.4, 30
In conclusion, modality of second-line treatment of epithelial ovarian carcinoma should be determined more by assessment of performance status than age. Second-line treatment with topotecan or paclitaxel-carboplatin can be administered safely in the aged. The elderly thus should benefit the full extent from recent advances in second-line treatment of epithelial ovarian carcinoma.