What's New in Predict

April 2017
  • A validation of PREDICT V1.2 in patients under 50 years of age at diagnosis is published in European Journal of Cancer.
  • Engelhardt EG, van den Broek AJ, Linn SC, et al. Accuracy of the online prognostication tools PREDICT and Adjuvant! for early-stage breast cancer patients younger than 50 years. Eur J Cancer 2017;78:37-44.
  • A paper published recently in the JNCI reports on a new breast cancer prognostic model. The authors state in the abstract that “Our model further showed better calibration compared with PREDICT”. However, this statement is not backed up by the results presented in the Supplementary Table 3. In the validation data set in which the new model and PREDICT were compared the observed number of deaths at five years was 148 compared to 116 predicted by their full model (P-difference = 0.003), 109 predicted by their clinical model (P = 0.0002) and 128 predicted by PREDICT (P = 0.08). Far from showing better calibration than PREDICT, it shows worse calibration. The authors do not report on the comparative performance of the two models based on discrimination.
  • Wu X, Ye Y, Barcenas CH, et al. Personalized Prognostic Prediction Models for Breast Cancer Recurrence and Survival Incorporating Multidimensional Data. J Natl Cancer Inst 2017;109(7).
November 2016
  • PREDICT has been endorsed by the American Joint Committee on Cancer in the AJCC Cancer Staging Manual, 8th edition.
  • The AJCC Precision Medicine Core (PMC) developed and published criteria for critical evaluation of prognostic tool quality1 which are presented and discussed in Chapter 4 of the Manual. Although developed independently by the PMC, the AJCC quality criteria correspond fully with the recently developed Cochrane CHARMS CHecklist for critical Appraisal and data extraction for systematic Reviews of prediction Modeling Studies 2.
  • 1. Kattan MW, Hess KR, Amin MB, et al. American Joint Committee on Cancer acceptance criteria for inclusion of risk models for individualized prognosis in the practice of precision medicine. CA: a cancer journal for clinicians. 2016.
  • 2. Moons KG, de Groot JA, Bouwmeester W, et al. Critical appraisal and data extraction for systematic reviews of prediction modelling studies: the CHARMS checklist. PLoS medicine. 2014;11(10):e1001744.
August 2016
  • The overall fit of the underlying model has been good in multiple independent case series, but PREDICT has been shown to underestimate breast cancer specific mortality in women diagnosed under the age of 40. Another limitation of the model is the use of discrete categories for tumour size and node status which result in “step” changes in risk estimates on moving from one category to the next. We have therefore refitted the PREDICT prognostic model using the original cohort of cases from East Anglia with updated survival time in order to take into account age at diagnosis and to smooth out the survival function for tumour size and node status.
  • Multivariable Cox regression models were used to fit separate models for ER negative and ER positive disease. Continuous variables were fitted using fractional polynomials. The fit of the prognostic models were then tested in three independent data sets that had also been used to validate the original version of PREDICT.
  • We have also modified the benefits of adjuvant chemotherapy applied in the model. The benefits applied in PREDICT v1 were the same as those used by the Adjuvant!Online model and were derived from the Early Breast Cancer Trialists Collaborative Group 2005 paper. The update of the EBCTCG analysis of polychemotherapy trials published in 2012 (Early Breast Cancer Trialists Collaborative Group. Comparisons between different polychemotherapy regimens for early breast cancer: meta-analyses of long-term outcome among 100,000 women in 123 randomised trials. Lancet 2012;379(9814):432-44.) reported that "For anthracycline- based regimens, there was no good evidence of any heterogeneity of the proportional risk reductions with age, nodal status, ER status, tumour differentiation, tumour diameter, or combinations of these.” We have therefore applied a single relative risk reduction of 22 per cent for all patients for standard anthracycline based regimes and 36 per cent for taxane based regimes.
  • The overall calibration and discrimination of the new version of PREDICT (v2.0) was good and comparable to that of the previous version in both model development and validation data sets. However, the calibration of v2.0 improved over v1.2 in patients diagnosed under the age of 40.
  • A manuscript reporting these results has been submitted. Until this has been accepted for publication PREDICT v2.0 will be availabe as a beta-version.
  • An option to select micrometastases has been added for node negative patients. The prognosis for these cases is treated as intermediate between node negative and a single positive node.
April 2016
  • PREDICT discussed in a review of internet tools to enhance breast cancer care. Shachar SS, Muss HY. Internet tools to enhance breast cancer care. npj Breast Cancer 2016;2. (doi:10.1038/npjbcancer.2016.11).
January 2015
  • Unnecessary text on PDF/print output has now been removed. Additionally a note for Internet Explorer (IE) users; to facilitate robust PDF/print output please update your PDFCreator software to version 2.0. IE should prompt for this update when PDF/print output is requested.
November 2014
  • The accuracy of PREDICT with and without KI67 has been evaluated using a case-cohort from Nottingham. The addition of KI67 led to a small imporvement in calibration and discimination in 1,274 patients with ER+ disease - the area under the ROC curve improved from 0.7611 to 0.7676 (p=0.005). These data have been accepted for publication in BMC Cancer. The potential impact of PREDICT on chemotherapy/trastuzumab recommendations in HER2‐positive patients with early‐stage breast cancer has recently been evaluated in 109 patients who presented to the Cambridge Breast Unit (SK Down, O Lucas, JR Benson, GC Wishart. Effect of PREDICT on chemotherapy/trastuzumab recommendations in HER2-positive patients with early-stage breast cancer. Oncol Lett 2014;8(6):2757-2761.) This study showed that PREDICT can aid decision‐making in HER2‐positive early-stage breast cancer by identifying older patients at risk of undertreatment with chemotherapy/trastuzumab, and by reducing the overtreatment of patients with little predicted benefit, particularly in ER‐positive disease.
  • Predict Graphics chart code is now developed with the google jsapi (Javascript API). Barchart values may be viewed by hovering over the bar, or tapping bar on a mobile device.
  • Predict cannot now be guaranteed to work on older browsers such as Internet Explorer (IE) pre version 7, and Mozilla Firefox pre version 31.
  • Updated information and policy regarding cookie usage, see Privacy Policy.