MedWire News: IBTR! version 2.0, a web-based predictive nomogram for ipsilateral breast tumor recurrence (IBTR) after breast conserving therapy (BCT), accurately predicts IBTR risk in most low-to-moderate risk patients but still overestimates risk in a minority of patients with higher risk features, research shows.
IBTR! version 1.0 was developed by Alphonse Taghian (Massachusetts General Hospital, Boston, USA) and colleagues to estimate individualized risk for IBTR after BCT but preliminary testing demonstrated over-estimation in high-risk subgroups.
This led the researchers to modify and retest the model using two independent population-based datasets. They performed Cox regression modeling on 7811 patients treated with BCT (median follow-up, 9.4 years) to generate population-based hazard ratios for each of the seven clinicopathologic variables in the IBTR! nomogram. These revised hazard ratios formed the basis for the modified IBTR! version 2.0.
Taghian and team then tested IBTR! version 2.0 against a further 664 patients (median follow-up, 9.3 years). They compared the mean predicted and observed 10-year estimates for the entire cohort and for four groups predefined by nomogram-predicted risks: group 1: less than 3%; group 2: 3% to 5%; group 3: 5% to 10%; and group 4: more than 10%.
The researchers report that, overall, there were only 22 cases of IBTR as a first event in the validation data set. IBTR! version 2.0 predicted an overall 10-year IBTR estimate of 4.0%, while the observed estimate was 2.8%. The predicted and observed IBTR estimates were 2.2% versus 1.3% in group 1 (n = 283), 3.8% versus 3.5% in group 2 (n = 237), 6.7% versus 3.2% in group 3 (n = 111), and 12.5% versus 8.7% group 4 (n = 33).
“The differences between predicted and observed IBTR estimates in groups 3 and 4 are difficult to interpret,” remark Taghian and co-authors in the Journal of Clinical Oncology. They note that the differences were not statistically significant but admit the IBTR event rates in these groups were relatively low, and the study could therefore be under-powered.
The team concludes: “Validation testing of this promising tool in a larger prospective data set is warranted.”
MedWire (www.medwire-news.md) is an independent clinical news service provided by Current Medicine Group, a trading division of Springer Healthcare Limited. © Springer Healthcare Ltd; 2010
Free abstract
