Nomogram for Early Mortality in Metastatic Colorectal Cancer


Key Points

  • A nomogram was developed that predicted early mortality in metastatic colorectal cancer with strong discrimination.
  • The model for 90-day mortality includes increased age, decreased body mass index, worse performance status, BRAF-mutant status, increased number of metastatic sites, and increased bilirubin, white blood cell count, and absolute neutrophil count.

As reported in the Journal of Clinical Oncology, Renfro et al have derived a nomogram for predicting early mortality in metastatic colorectal cancer using data from more than 20,000 patients in the ARCAD (Aide et Recherche en Cancérologie Digestive) database.

Study Details

The analysis involved pooled data from 22,654 patients from 28 phase III trials in the ARCAD database. Multivariable logistic regression models for 30-, 60-, and 90-day mortality were constructed, including clinically and statistically significant patient and disease factors and interaction terms. A nomogram for 90-day mortality was developed and validated internally in the construction dataset (n = 20,397) and then externally using a 10% random holdout sample from each trial (n = 2,230).

Models for Early Mortality

In the construction set, mortality rates were 1.4% at 30 days, 3.4% at 60 days, and 5.5% at 90 days. The final multivariable model for mortality by 30 days included increased age (P < .001), decreased body mass index (P < .001), worse Eastern Cooperative Oncology Group/World Health Organization performance status (P < .001), BRAF-mutant status (P < .001), increased number of metastatic sites (P < .001), increased bilirubin (P < .001), increased white blood cell count (P < .001), and increased derived neutrophil-to-lymphocyte ratio (P = .0046). The final models for 60- and 90-day mortality replaced the derived neutrophil-to-lymphocyte ratio with increased absolute neutrophil count (P < .001 in both models). The final model for 90-day mortality exhibited strong internal discrimination, with a C-index of 0.772, and good discrimination across patient subgroups.

External Validation

In the external validation set, 90-day mortality was 5.7% (95% confidence interval [CI] = 4.8%–6.8%), with the final model-predicted rate of 5.0% falling within the 95% confidence interval of the actual rate. The final model-predicted rates of 90-day mortality also fell within the 95% confidence intervals of all patient subgroups defined by levels of each variable in the final model, except for the subgroup of patients with ≥ 4 metastatic sites (predicted rate of 8.7%; actual rate of 4.0%, 95% CI = 2.3%–6.8%).

The investigators concluded: “A validated clinical nomogram has been developed to quantify the risk of early death for individual patients during initial treatment of metastatic colorectal cancer. This tool may be used for patient eligibility assessment or risk stratification in future clinical trials and to identify patients requiring more or less aggressive therapy and additional supportive measures during and after treatment.”

Lindsay A. Renfro, PhD, of the Mayo Clinic, Rochester, is the corresponding author of the Journal of Clinical Oncology article.

The study was supported by the National Center for Advancing Translational Science of the National Institutes of Health and the ARCAD database.

The content in this post has not been reviewed by the American Society of Clinical Oncology, Inc. (ASCO®) and does not necessarily reflect the ideas and opinions of ASCO®.