PLCOm2012 Model May Be More Effective in Screening African Americans for Lung Cancer

A study of the PLCOm2012 model, published in the Journal of Thoracic Oncology, found the model to be superior to the United States Preventive Services Task Force (USPSTF) criteria at identifying African American ever-smokers for lung cancer screening.¹

Given these findings, researchers indicated that eligibility criteria for lung cancer screening guidelines should be modified to include the use of a more accurate lung cancer predication model, such as that outlined in the current study, to avoid worsening the observed disparities between white and African American ever-smokers in lung cancer outcomes.

“We found that the PLCOm2012 model was significantly more sensitive in selecting lung cancer patients as being eligible for screening for the entire cohort,” lead author Mary Pasquinelli, DNP, from the University of Illinois at Chicago, said in a press release.² “Broader use of this model in racially diverse populations may help overcome disparities in lung cancer screening and outcomes.”

The PLCOm2012 is a validated logistic regression lung cancer risk prediction model based on data collected from the control arm of the Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial (PLCO), which is a randomized controlled trial studying screening to reduce cancer mortality. The model incorporates 11 predictors, including sociodemographic, medical history, and 4 smoking exposure variables.

In this retrospective review of lung cancer screening eligibility based on the USPSTF criteria versus the PLCOm2012 model, investigators evaluated lung cancer cases diagnosed between 2010 and 2019 at an urban medical center which serves a racially and ethnically diverse population. Overall, 883 ever-smokers were identified, including 258 (29.2%) white, 497 (56.3%) African American, 69 (7.8%) Hispanic), 24 Asian(2.7%), and 35 (4.0%) other.

Compared with the USPSTF criteria, the PLCOm2012 model was found to increase sensitivity in for the African American cohort lung cancer risk thresholds of 1.51%, 1.70%, and 2.00% per 6-years (P < .0001). For instance, at the 1.70% risk threshold the PLCOm2012 model identified 71.3% African American cases, while the USPSTF criteria comparatively identified 50.3% (P < .0001). Contrastingly, there was no difference observed in white cases (66.0% vs 62.4%, respectively [P = .203]).

“The current study demonstrates that the sensitivity to detect lung cancers overall is lower for the USPSTF criteria than for the PLCOm2012 model at all three risk thresholds studied,” the authors wrote. “Moreover, the USPSTF criteria significantly under-selects African American compared to white ever-smokers.”

Moreover, across the entire study cohort, the PLCOm2012 model at the 1.7% threshold was more sensitive than USPSTF in screening for lung cancer. Of the 883 identified cases, only 44 (5.0%) were USPSTF criteria positive and PLCOm2012 1.7% risk or greater threshold negative, whereas 166 (18.8%) were PLCOm2012 1.7% risk or greater threshold positive and USPSTF negative (McNemar’s odd ratio = 3.77; 95% CI, 2.69-5.39; P < .0001).

“Enough evidence has accumulated that we know with certainty that disparities exist in selecting African American ever-smokers for lung cancer screening using the USPSTF or [Center for Medicare and Medicaid Services; CMS] criteria,” the authors wrote. “The results of this historical cohort study using the PLCOm2012 model have implications for public and clinical health.”

Moving forward, work is currently underway to develop smartphone and tablet apps which include the PLCOm2012 model and can be used in clinics or by individuals at home or elsewhere. Further, a pilot study is already underway to evaluate patient self-completion and calculation of PLCOm2012 score using tablets in doctors’ offices before meeting with clinicians.

“Such products should make PLCOm2012 risk assessment easier, more flexible and reduce assessment time,” the authors wrote.

References:

1. Pasquinelli MM, Tammemägi MC, Kovitz KL, et al. Risk Prediction Model versus United States Preventive Services Task Force Lung Cancer Screening Eligibility Criteria – Reducing Race Disparities. Journal of Thoracic Oncology. doi: 10.1016/j.jtho.2020.08.006

2. Current lung cancer public health screening guidelines under count African Americans [news release]. Denver. Published August 18, 2020. Accessed August 18, 2020. https://www.eurekalert.org/pub_releases/2020-08/iaft-clc081820.php