Model Provides Template for High-Income & Emerging Economies to Plan Cancer Surgery Provision

A guideline-based model described in The Lancet Oncology may provide a new template for high-income and emerging economies to rationally plan and evaluate their cancer surgery provision.

The model replicates examples of benchmarking radiotherapy and chemotherapy demand for cancer surgery across all major site-specific cancers. Investigators suggested the model could be used to identify service gaps at the population level after adjusting to the important local variations in cancer epidemiology.

“The model could also be used to estimate the future demand and the resource requirements for achieving country-specific benchmarks on cancer surgical service provision,” wrote the study authors, who were led by Sathira Kasun Perera, MSc.

In this study, investigators analyzed the latest clinical guidelines from high-income countries and charted surgical treatment pathways for 30 malignant cancer sites, including 19 individual sites and 11 grouped as other cancers, that were notifiable in Australia in 2014. Cancer stage, tumor characteristics, and fitness for surgery, and other population-based epidemiological data were also derived from Australia and other similar high-income settings for 2017.

The probabilities across the clinical pathways of each cancer were taken together to estimate the population-level benchmark rates of cancer surgery. These rates were then further validated by comparing the observed rates of cancer surgery in the South Western Sydney Local Health District from 2006 to 2012.

Overall, surgical treatment was found to be the most desirable treatment option at least once during the course of treatment for 58% of newly diagnosed patients with cancer in Australia (95% CI, 57%-59%) in 2014; however, this varied by site from 23% for prostate cancer (95% CI, 17%-27%) to 99% for testicular cancer (95% CI, 96%-99%). Cancer surgery rates in South Western Sydney were similar to the benchmarks for most cancers, though they were higher for some cancers, such as prostate (absolute increase of 29%), and lower for others, such as lung cancer (–14%).

The variability of benchmark surgery rate for each cancer site was mapped by the investigators against the range of values used for all input parameters. Ultimately, it was revealed that the models for all cancer sites detailed robustness of the outcome over the range of input values used.

Importantly though, this population-based analysis was not without limitations, including that the accuracy of the model-predicted benchmarks estimates is limited by the agreement of guideline recommendations. However, in an attempt to address this limitation, the investigators also performed univariable and multivariable sensitivity analyses to capture all relevant perspectives.

“This modeling exercise provides benchmarks relevant to populations and should not be used to assess individual institutions or provider performance because referral patterns are not representative of entire populations,” the authors noted. “For example, it is unlikely that unfit individuals will be referred for surgery. Nevertheless, the intention was to assist in identifying services gaps at the population level and predicting the current and future demand for cancer surgery.”

By using benchmark surgery rates to predict the cancer surgical demand, however, the investigators indicated that this model has the advantage of eliminating confounding supply-side limitations, such as gaps in accessibility, availability, and variation in referral patterns.

Reference:

Perera SK, Jacob S, Sullivan R, Barton M. Evidence-based benchmarks for use of cancer surgery in high-income countries: a population-based analysis. Lancet Oncol. 2021;22(2):173-181. doi: 10.1016/S1470-2045(20)30589-1