1 Service d'oncologie médicale, Centre Hospitalier Lyon Sud, Pierre-Bénite; 2 Service d'oncologie médicale, Hôtel Dieu, Paris; 3 Département d'information médicale, Hospices Civils de Lyon, Lyon; 4 Service d'oncologie médicale, Centre Hospitalier Universitaire, Grenoble; 5 Service d'oncologie médicale, Centre Alexis-Vautrin, Vandoeuvre-les-Nancy; 6 Institut cancérologique de la Loire, Saint-Etienne; 7 Service d'oncologie médicale, Clinique Jean Bernard, le Mans; 8 Service d'oncologie médicale, Hôpital universitaire Dupuytren, Limoges; 9 Service d'oncologie médicale, Hôpital d'Instruction des Armées Ste Anne, Toulon Naval, France
* Correspondence to: Dr G. Freyer, Medical Oncology Department, Centre Hospitalier Lyon Sud, 69495 Pierre-Bénite Cedex, France. Tel: +33-47-88-64-318; Fax: +33-47-88-64-319; E-mail: gilles.freyer{at}chu-lyon.fr
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Patients and methods: From July 1998 to October 2000, 83 advanced ovarian carcinoma patients >70 years old received carboplatin AUC 5 and cyclophosphamide 600 mg/m2, on day 1 of six 28-day cycles. The clinical and biological geriatric covariates prospectively studied were: comorbidities, comedications, cognitive functions (Mini-Mental test), nutritional status and autonomy.
Results: Patient characteristics were: median age 76 years, serous histology (73%), FIGO stage III (75%), optimal initial surgery (21%) and performance status (PS) 2 (44%). Sixty patients (72%) received six chemotherapy cycles without severe toxicity (STox) or tumor progression. Multivariate analysis retained three factors as independent predictors of STox: symptoms of depression at baseline (P = 0.006), dependence (P = 0.048) and PS
2 (P = 0.026). Independent prognostic factors identified for overall survival (Cox model) were depression (P = 0.003), FIGO stage IV (P = 0.007) and more than six different comedications per day (P = 0.043).
Conclusion: CGA could predict STox and overall survival of elderly advanced ovarian carcinoma patients.
Key words: chemotherapy, elderly, geriatric assessment, ovarian cancer
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The aim of this study, specifically performed in elderly patients with stage III/IV ovarian carcinoma, was to evaluate the ability of some CGA parameters to predict efficacy and tolerance of the carboplatin + cyclophosphamide combination, in order to identify which subgroups of elderly patients should receive standard or dose-reduced therapy and for whom chemotherapy would not be beneficial.
![]() |
Patients and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Patients
We included patients >70 years old with stage III or IV ovarian epithelial carcinoma, who met the following criteria: cytologically or histologically proven epithelial carcinoma, initial laparotomy or not, normal blood counts with neutrophils >1.5 g/l and platelets >100 g/l, no prior irradiation, and absence of icterus. A patient was eligible, regardless of PS (ECOG), if her treating physician considered that she could receive chemotherapy.
Pretreatment screening
Screening before inclusion in the study comprised: PS evaluation, complete blood counts, hemoglobin, aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase, -glutamyl transferase, blood electrolytes, serum creatinine and urea, CA-125 assay, biological nutritional status (see Comprehensive Geriatric Assessment), electrocardiogram, and computed tomography (CT) scans of the chest, abdomen and pelvis.
Treatment
Patients were scheduled to receive carboplatin AUC 5 according to the Chatelut formula + cyclophosphamide 600 mg/m2 every 28 days for six cycles. Carboplatin and cyclophosphamide were given during successive 30-min infusions. Antiemetics were given intravenously: ondansetron 8 mg or granisteron 3 mg + methylprednisolone 1 mg/kg. An intermediate laparotomy after three cycles was not planned.
Dose reductions were planned as follows: for grade 4 neutropenia lasting >7 days, febrile neutropenia, grade 4 thrombocytopenia, incomplete hematological recovery on day 35 (neutrophils <1.5 g/l, platelets <100 g/l) or grade 2 neurotoxicity, the carboplatin dose was reduced to AUC 4 and the cyclophosphamide dose to 500 mg/m2 for all subsequent cycles. If toxicity persisted despite dose reduction, chemotherapy was considered unfeasible.
Treatment feasibility was defined as the patient's ability to receive six courses of this chemotherapy regimen without disease progression, death due to any cause, patient's or investigator's decision to stop treatment, persistent toxicity as define above despite dose reduction, or any grade 3/4 non-hematological toxicity.
Comprehensive Geriatric Assessment
During the consultation immediately preceding inclusion and first chemotherapy cycle, each patient completed with the investigator the CGA evaluation. Patient autonomy was assessed according the following categories: full autonomy at home or dependence (living at home with assistance, living with medical assistance in a specialized institution). We also recorded all comorbidities, mainly heart and vascular diseases, respiratory disease, diabetes, liver and kidney function test results, and the number of different drugs taken daily at baseline (zero to three, four to six, or more than six).
Nutritional status was assessed by body mass index (BMI), protidemia, albuminemia and total cholesterol. Cognitive function was evaluated with the Mini-Mental Status (MMS) test developed by Folstein et al. [9]. The presence or absence of clinical symptoms of depression was assessed by the investigator.
Evaluation criteria for toxicity and efficacy
Chemotherapy-induced side-effects were assessed at each cycle using the National Cancer Institute Common Toxicity Criteria, version 1. We defined the STox as the occurrence of at least one of the following events: febrile (>38.5°C) neutropenia, grade 4 neutropenia, early treatment withdrawal because of grade 3/4 toxicity or re-hospitalization for >7 days for grade 3/4 toxicity.
For patients with initial measurable tumor masses >2 cm in the largest diameter, tumor response was evaluated by CT scans after three and six cycles of chemotherapy.
Statistical analysis
The number of patients to be included was calculated with the intent to reject the null hypothesis of a <59% rate of chemotherapy feasibility, with = 10% and ß = 5% (unilateral hypothesis). Based on this assumption, 80 patients should be included.
Survival analysis
Overall survival, from inclusion to the patient's death, and progression-free survival, from study entry until disease progression confirmed clinically, radiologically or biologically (CA-125 increase), were estimated using the KaplanMeier method with censored data and compared using the log-rank test.
Predictive STox and prognostic factors
To predict STox and identify prognostic factors, we analyzed the occurrence of chemotherapy-induced STox as defined above. Progression-free and overall survival and CGA parameters were subjected to univariate analyses using the Pearson 2-test and Fisher's exact tests for categorical variables, and the non-parametric MannWhitney U-test for continuous variables. The following covariates were assessed at baseline: (i) chronological age; (ii) PS; (iii) MMS; (iv) autonomy; (v) comorbidities; (vi) number of daily comedications; (vii) nutritional status; (viii) hematological, renal and liver function parameters; (ix) tumor characteristics: FIGO stage (III versus IV), histology (serous subtype versus other), optimal initial surgery with residual tumor masses <1 cm (versus non-optimal surgery), absence (versus presence) of ascites or pleural effusion, involvement of liver or lung (versus no involvement), number of involved sites, and CA-125 level.
A logistic regression model was constructed by stepwise introduction of covariates that had been significantly associated with STox criteria (P < 0.05), according to our univariate analysis.
To identify the prognostic factors of progression-free and overall survival, a Cox model was used in which significant covariates coming from the univariate analysis (P < 0.05), were introduced stepwise to obtain the final model.
All tests were two-sided. Analyses were performed using the SPSS® version 10.0 statistical package (SPSS Inc., Chicago, IL, USA).
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
The toxicities observed are reported in Table 3 as the percentages of patients affected. Concerning hematological toxicities, grade 3/4 neutropenia, anemia and thrombocytopenia were observed in 28.7%, 9% and 8.3% of the cycles, respectively.
|
Twenty-three patients (27.7%; 95% CI 19.9% to 39.3%) experienced STox: six developed febrile neutropenia, six discontinued treatment because of hematological toxicity, one grade 3 mucositis, one grade 3 asthenia, one death consecutive to cranial trauma, and hospitalization lasting >7 days for one patient with hematological toxicity, four patients with grade 3 infection, one with grade 4 infection, one with femoral trauma and one with thrombophlebitis. According to our univariate analysis, the covariates reaching statistical significance were: PS 2 (P = 0.007), dependence (P = 0.017) and symptoms of depression at baseline (P = 0.006). FIGO stage IV tended towards significance (P = 0.075).
The multivariate analysis retained the following independent factors as predictive of STox: depression (P = 0.006), PS 2 (P = 0.026) and dependance (P = 0.048).
Efficacy
Data on CA-125 evolution during chemotherapy were available for 64% of patients. The CA-125 level normalized in 31.4% and decreased <50% in the remaining 24.4%. The radiological tumor response rate, assessable in 48 patients with initial measurable masses, was 48%. Fourteen patients underwent a second-look laparotomy: one histological complete response, three macroscopic complete responses with persistent of histological involvement and six partial responses with macroscopic residual disease were observed.
Survival
Median progression-free survival for the entire population was 9.9 months (95% CI 7.212.6) (Figure 1A). The multivariate analysis retained the following independent prognostic factors as being independently associated with poorer progression-free survival: depression (P<0.003), FIGO stage IV (P <0.04) and initial non-optimal surgery (P <0.008).
|
Figure 1C shows overall survival curves for patients with zero, one, or two or more poor prognosis factors.
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The eligibility criteria for our prospective study, which enrolled 83 elderly patients with advanced ovarian carcinoma, were not restrictive, as we included women with poor PS (3 and 4) to avoid selection bias. However, since there was no systematic registration of non-included patients, our results cannot be generalized to the elderly population. The patients of our study had relatively few geriatric conditions (Table 2) and an average life expectancy of 22 months. Despite the motivation of the investigators, more fragile patients were not included. Such patients, usually living in a geriatric institution, are probably not referred to an oncologist.
In 1998, GINECO decided to use the carboplatin + cyclophosphamide combination because of its reported good tolerance and efficacy [15, 16
]. Knowing that the International Collaborative Ovarian Neoplasm group recently described similar efficacies of carboplatin alone versus its combination with paclitaxel [17
], we can retrospectively admit that carboplatin alone was a reasonable alternative.
However, in our study population, among which 44% of patients had PS 2 and 25% had FIGO stage IV, the carboplatincyclophosphamide combination was well tolerated, since 72% of patients received six cycles without STox or disease progression. This is an encouraging observation, which leads us to recommend that this combined regimen be a standard in the population of women over 70 years of age.
Furthermore, we must emphasize that initial surgery had been optimal in only 20% of our patients. This is not surprising, since very similar data have already been reported in epidemiological studies [1]. In our opinion, given the clinical consequences of incomplete initial laparotomy in elderly patients with peritoneal carcinomatosis extending beyond the pelvis, and often ascites, chemotherapy as first-line treatment should always be considered.
Among patients who experienced STox necessitating treatment discontinuation, some variables appear to have predictive value. One factor, PS, is not specific to elderly patients. Indeed, it is a variable that is usually associated with patients' prognosis and/or treatment tolerance in most studies examining various tumor types. Notably, depression symptoms at baseline and loss of autonomy, variables included in the CGA, were predictive of chemotherapy-induced STox. STox is a composite parameter that takes into account all deleterious clinical events in elderly patients: re-hospitalization for STox lasting >7 days, febrile neutropenia and early treatment withdrawal because of toxicity. This parameter must be distinguished from chemotherapy feasibility, which also includes absence of tumor progression and thus could not be included in the predictive analysis.
Symptomatic depression assessment was evaluated by the investigator, who did not use any specific scale or questionnaire. Although the pertinence of this assessment may be questionable from a psychiatric point of view, this parameter, easily assessable, simple and pragmatic, seems valuable from a statistical point of view and is independent of PS and cognitive status as assessed by the MMS. In a recent study, Audisio et al. [18] showed a correlation between Geriatric Depression Scale (GDS) score and surgical morbidity. The use of GDS, a simple and validated tool to assess depression in the elderly, should certainly be recommended in further predictive studies.
The patients' degree of autonomy, which is not directly linked to the PS, was also an independent predictor of STox. Using the Instrumental Activities of Daily Living score [19] to evaluate the autonomy of patients with hematological malignancies, one study showed this parameter to have a prognostic value for survival [20
].
Concerning overall survival, we identified three baseline prognostic factors: disease stage is routinely identified but not specifically associated with older age, while retention of clinical symptoms of depression and comedications (more than six drugs per day) confirmed the relevance of our approach. Comedication reflects comorbidity: the more drugs a patient takes, the more concomitant diseases. Frasci [21] underlined the contribution of the evaluation of comorbidities for predicting chemotherapy toxicity in lung cancer patients.
We must emphasize that chronological age, which was included in the Cox model, did not appear to have a prognostic value, in either the univariate or multivariate analysis.
Nevertheless, a potential methodological limitation in this study, other than possible patient-selection bias, was the post-chemotherapy surgical debulking performed in 14 patients, which was not taken into account. We think that this procedure did not dramatically modify our analysis of potential prognostic factors, because only responding patients, thus with better prognoses, underwent subsequent laparotomy.
Our results demonstrate that some simple parameters that can be systematically assessed in each patients >70 years old may help the physician choose the best therapeutic strategy. For patients with two or three factors predictive of poor prognosis, the use of monochemotherapy (carboplatin) should probably be chosen, in order to minimize the likelihood of STox. Best supportive care, without any specific anticancer treatment, should also be considered.
Among the limits of our methodology, the use of a possibly suboptimal chemotherapy and the lack of standardized geriatric evaluation based on validated scales [7, 8
] must be stressed.
As a consequence, GINECO has undertaken another prospective study, using the same methodology, for elderly women with advanced ovarian carcinoma, but changing the chemotherapy regimen to paclitaxel + carboplatin and conducting a more exhaustive CGA using the Instrumental Activities of Daily Living, the timed up-and-go test [22], the Prognostic Inflammatory and Nutritional Index (PINI) given by the ratio albumin x pre-albumin/C-reactive protein x orosomucoid [23
], and the Hospital Anxiety and Depression Scale. The timed up-and-go test allows the investigator to assess the patient's physical capacities in addition to autonomy and the PINI seems to provide a more accurate assessment of the patient's nutritional status, since the biological parameters tested in our study had no evident prognostic value.
We intend to validate the usefulness of our analytical stategy to identify prognostic factors with the data being generated by the ongoing study. The coadministration of paclitaxel will be included in this analysis as a potentially prognostic variable.
In our opinion, prospective trials on elderly patients should take into account potentially prognostic items included in the CGA. As emphasized by Extermann et al. [24], some prognostic indexes need to be established to help physicians treat elderly patients. The value of those indexes could be at least equal or superior to those of the usual, clinical and biological prognostic factors used in routine oncology practice.
![]() |
Acknowledgements |
---|
Received for publication April 20, 2005. Revision received July 12, 2005. Accepted for publication July 12, 2005.
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2. Alberts DS, Dahlberg S, Green SJ et al. Analysis of patient age as an independent prognostic factor for survival in a phase III study of cisplatincyclophosphamide versus carboplatincyclophosphamide in stages III (suboptimal) and IV ovarian cancer. Cancer 1993; 71: 618627.[ISI][Medline]
3. Freyer G, Maire P, Ardiet C et al. Cancer chemotherapy in the elderly: present and future. Bull Cancer 1995; 82: 531540.[ISI][Medline]
4. Sargent D, Goldberg R, Jacobson S et al. A pooled analysis of adjuvant chemotherapy for resected colon cancer in elderly patients. N Engl J Med 2001; 15: 344345.
5. McKenna RJ. Clinical aspects of cancer in the elderly. Cancer 1994; 74: 21072117.[ISI][Medline]
6. Freyer G, Lortholary A, Delcambre C et al. Unexpected toxicities in elderly patients treated with oral idarubicin in metastatic breast cancer: the GINECO experience. Clin Oncol 2004; 16: 1723.[CrossRef][ISI]
7. Repetto L, Fratino L, Audisio RA et al. Comprehensive geriatric assessment adds information to Eastern Cooperative Oncology Group performance status in elderly cancer patients: an Italian Group for Geriatric Oncology Study. J Clin Oncol 2002; 20: 494502.
8. Monfardini S, Balducci L. A comprehensive geriatric assessment (CGA) is necessary for the study and the management of cancer in the elderly. Eur J Cancer 1999; 35: 17711772.[CrossRef][ISI][Medline]
9. Folstein MF, Folstein SE, McHugh PR. "Mini-mental status". A practical method for grading the cognitive state of patients for the clinician. J Psychiat Res 1975; 12: 189198.[CrossRef][ISI][Medline]
10. Mc Guire WP, Hoskins WJ, Brady MF et al. Cyclophosphamide and cisplatin compared with paclitaxel and cisplatin in patients with stage III and stage IV ovarian cancer. N Engl J Med 1996; 334: 16.
11. Kemeny MM, Peterson BL, Kornblith AB et al. Barriers to clinical trial participation by older women with breast cancer. J Clin Oncol 2003; 21: 22682275.
12. Gershenson DM, Follen Mitchell M, Atkinson N et al. Age contrasts in patients with advanced epithelial ovarian cancer: the MD Anderson Cancer Center experience. Cancer 1993; 71: 638643.[ISI][Medline]
13. Castiglione M, Gelber RD, Goldhirsch A. Adjuvant systemic therapy for breast cancer in the elderly: competing causes of mortality. J Clin Oncol 1990; 8: 519526.
14. Ibrahim NK, Frye DK, Buzdar AU et al. Doxorubicin-based chemotherapy in elderly patients with metastatic breast cancer. Tolerance and outcome. Arch Intern Med 1996; 156: 882888.[Abstract]
15. Swenerton K, Jeffrey J, Stuart G et al. Cisplatincyclophosphamide versus carboplatincyclophosphamide in advanced ovarian cancer: a randomized phase III study of the National Cancer Institute of Canada Clinical Trials Group. J Clin Oncol 1992; 10: 718726.
16. Hannigan EV, Green S, Alberts DS et al. Results of a Southwest Oncology Group phase III trial of carboplatin plus cyclophosphamide versus cisplatin plus cyclophosphamide in advanced ovarian cancer. Oncology 1993; 50: 29.[ISI][Medline]
17. International Collaborative Ovarian Neoplasm Group. Paclitaxel plus carboplatin versus standard chemotherapy with either single-agent carboplatin or cyclophosphamide, doxorubicin, and cisplatin in women with ovarian cancer: the ICON3 randomised trial. Lancet 2002; 360: 505515.[CrossRef][ISI][Medline]
18. Audisio RA, Ramesh H, Pope DP et al. Preoperative assessment of cancer in the elderly: interim analysis. Proc Am Soc Clin Oncol 2005; 736s: 8028.
19. Lawton MP, Moss M, Fulcomer M et al. A research and service oriented multilevel assessment instrument. J Gerontol 1982; 37: 9199.[ISI][Medline]
20. Zagonel V. Importance of a comprehensive geriatric assessment in older cancer patients. Eur J Cancer 2001; 37 (Suppl 7): S229S233.[ISI][Medline]
21. Frasci M. Chemotherapy of lung cancer in the elderly. Crit Rev Oncol Hematol 2002; 41: 349361.[ISI][Medline]
22. Podsialo D, Richardson S. The timed "up and go": a test of basic functional mobility for frail elderly persons. J Geriatr Soc 1991; 39: 142148.[ISI]
23. Bonnefoy M, Ayzac L, Ingenbleek Y et al. Usefulness of the prognostic inflammatory and nutritional index (PINI) in hospitalized elderly patients. Int J Vitam Nutr Res 1998; 68: 189195.[ISI][Medline]
24. Extermann M, Chen H, Cantor AB et al. Predictors of tolerance to chemotherapy in older cancer patients: a prospective pilot study. Eur J Cancer 2002; 38: 14661473.[CrossRef][ISI][Medline]
|