Background
Patients with recurrent ovarian cancer are defined as \'platinum sensitive\' based on a treatment free interval of more than 6 months following prior platinum chemotherapy, but are a heterogeneous group with variable response to treatment and an unpredictable survival. We
developed and validated a nomogram to improve prediction of survival in these patients treated with platinum-based chemotherapy.
Methods
The nomogram was developed in a training cohort (n=934) from the CALYPSO trial and
validated in the AGO-OVAR 2.5 study (n=339). The proportional-hazards model (nomogram)
was based on pre-treatment characteristics. Patients were classified into good, intermediate and poor prognostic groups.
Results
The nomogram had a concordance index (c-index) of 0.673. Significant predictors ranked
according to importance were: performance status (20 points), tumour size (19 points), platinum chemotherapy-free interval (21 points), CA-125 (19 points), haemoglobin (14 points) and number of organ metastatic sites (7 points). The median survival for good, intermediate and poor prognosis was 56, 31, and 21 months respectively. When the nomogram was applied without CA-125 (CA-125 was not available in validation cohort), the c-indices were 0.653 (training) and 0.645 (validation). The calibration plot in the validation cohort based on five predictors (without CA-125)suggested good agreement between actual and nomogram-predicted 24-month survival
probabilities.
Conclusions
This nomogram, using six characteristics, improves survival prediction in patients with \'platinum sensitive\' recurrent ovarian cancer treated with platinum-based chemotherapy. It could lead to a more rational design and stratification according to risk in clinical trials and may be helpful in counselling patients prior to chemotherapy.
USER ACKNOWLEDGES AND AGREES THAT ALL DECISIONS MADE WITH THE ASSISTANCE OR USE OF THE SOFTWARE AND/OR THE WEBSITE AND/OR BASED ON CONTENT FOUND HEREIN WILL BE EXCLUSIVELY THE RESPONSIBILITY OF THE USER.
Powered by MULTIWEBCAST Technologies. MULTIWEBCAST PLATFORM Version 4.1000. All rights reserved.
