Predictive factors for chemotherapy feasibility in elderly patients with solid tumor: Results of GERCOR old prospective multicenter study

Reporter: Saumil Gandhi, MD PhD
The Abramson Cancer Center at the University of Pennsylvania
Last Modified: June 10, 2013

Presenter: Elisabeth Carola, MD
Presenter's Affiliation: University of Pittsburgh


  • It is estimated that one quarter of all patients with cancer are over the age of 75 years.
  • Many clinical trials suggest that tolerance of chemotherapy is unrelated to age. However, in clinical practice, older patients are treated less aggressively and often receive lower doses of chemotherapeutic drugs due to fear of toxicities.
  • The aim of this study is to identify predictors of chemotherapy feasibility in the geriatric population.
  • Identifying these predictors of chemotherapy feasibility would allow oncologists to optimize their treatment plans and improve the management of older patients with cancer.


  • This is a prospective multicenter cohort study.
  • Inclusion criteria were: Age ? 75 years, solid tumor, chemo-naive patient, and ability to receive at least 2/3 of the standard dose at the first course of treatment.
  • Ten geriatric parameters were recorded at baseline by the oncologist to assess cognitive function, dependence and mobility, co-morbidities, biologic parameters, and psychological and social environment.
    • 1-three words test
    • 2-date and address for cognitive function
    • 3-instrumental activities of daily living
    • 4- monopodal stand-up test
    • 5-hospitalization during the previous year
    • 6-number of medicines taken for comorbidities
    • 7-creatinine clearance
    • 8-albumin serum level
    • 9-self-rated depressive mood question
    • 10-presence of a caregiver
  • A multivariate logistic regression was used to build a predictive model.
  • The primary outcome was chemotherapy feasibility, which was defined as the ability to receive at least 3 months of the planned therapy.


  • 576 patients were enrolled at 49 different centers from 2008 to 2012
  • 516 (89.6%) were eligible for analysis.
  • Median age was 81 years with equal numbers of men and women. 69.5% of the patients were considered advanced stage and 74% had performance status 0-1.
  • 198 patients (38%) received monotherapy while 301 patients (58%) received doublet therapy.
  • FOLFOX, Fluoropyrimidine monotherapy, Gemcitabine monotherapy, and Paclitaxel-Carboplatin were the most commonly used regimens.
  • Chemotherapy feasibility was observed in 298 (57.8%) patients.
  • Although the majority of patients (82%) had performance status of 0 or 1, almost 40% of the patients presented with impaired mobility, significant comorbidities, and were found to be chemo unfeasible.
  • Grade 2+ toxicity was increased by 67% in chemo feasible patients versus chemo unfeasible patients (p=0.009).
  • On multivariate logistic regression analysis, only 2 geriatric parameters were significantly associated with chemotherapy unfeasibility: Albuminemia < 30g/l (adjusted OR = 2.34 CI95% [1.43-3.83]) and depressive mood (adjusted OR = 1.55 CI95% [1.02-2.35]).
  • Other geriatric parameters did not predict for chemotherapy feasibility.

Author's Conclusions

  • Albuminemia and self-rated depressive mood status are independent predictors of chemotherapy feasibility in elderly patients with solid tumors.
  • Unexpectedly, other geriatrics parameters were not independent predictors.

Clinical Implications

  • The GERCOR OLD study is one of the largest prospective studies performed in chemo-naive elderly patients.
  • Although a majority of patients (80%) were classified by oncologists as "fit" with good performance status, chemotherapy was feasible in only 60% of the patients. These findings highlight the inadequacy of clinical performance status in predicting the tolerance to chemotherapy.
  • Surprisingly, factors such as renal function (creatinine clearance) and medical comorbidities seem not to matter for predicting chemotherapy feasibility. The authors do not comment on why albumin levels and depressed mood alone determine a patient's tolerance for chemotherapy. It is possible that a patient's albumin levels are a general reflection of their overall health.
  • 20% of Americans are expected to be over the age of 65 by the year 2030. As our population continues to age, studies that provide predictive models to guide optimal therapy will become increasingly important.