Virtual Library
Start Your Search
Sukhinder Atkar-Khattra
Author of
-
+
FP09 - Screening and Early Detection (ID 175)
- Event: WCLC 2020
- Type: Posters (Featured)
- Track: Screening and Early Detection
- Presentations: 1
- Moderators:
- Coordinates: 1/28/2021, 00:00 - 00:00, ePoster Hall
-
+
FP09.01 - Economic Impact of Screening Selection with the PLCOm2012 Risk Model Versus USPSTF-Guidelines in the International Lung Screening Trial (ILST) (ID 2400)
00:00 - 00:00 | Author(s): Sukhinder Atkar-Khattra
- Abstract
Introduction
Screening protocols that use model-estimated risk for selection are thought to improve the efficiency of lung cancer screening by reducing the number needed to screen and detecting more lung cancers. Since risk selection enriches screening populations with greater age and tobacco smoke exposure, some hypothesize that screening with model-based risk selection would be less cost-effective than more broadly defined selection criteria such as those defined by the USPSTF. The objective of this study is to examine the costs and cost-effectiveness of selection with the PLCOm2012 risk model versus USPSTF-based selection in the International Lung Screening Trial (ILST).
Methods
The International Lung Screening Trial offered enrollment to participants that either had a PLCOm2012 lung cancer risk ≥1.51%/6 years or met the USPSTF criteria for screening selection. A cohort-based simulation model was developed to estimate the lifetime economic impact of screening PLCOm2012+ve but not PLCOm2012-ve/USPSTF+ve compared with screening USPSTF+ve but not PLCOm2012+ve/USPSTF-ve. We used risk-matched screening costs from the PanCan Early Detection of Lung Cancer study (PanCan) to calculate the post-baseline annual screening costs for each of the selection subgroups in the ILST and estimated the additional costs per quality adjusted life years (QALYs) with a published simulation model based on the NLST and PanCan screening studies.
Results
At the time of analysis 11,166 individuals registered in the study, with an interest in participating in lung cancer screening at one of seven coordinating sites in Canada, Australia and Hong Kong. Of these, 4282 enrolled in the study; 991 (23.1%) were exclusively selected by the PLCOm2012 criteria, 597 (13.9%) were exclusively selected by the UPSPSTF criteria and 2694 (62.9%) were selected by both criteria. The annual screening costs for the USPSTF-exclusive group were slightly higher than the rest of the cohort ($514 versus $508; p<0.05) and were thus held constant for simulation of the base-case scenario in the cost-effectiveness analysis. Compared with the PLCOm2012-selected cohort, the risk ratios for developing lung cancer were significantly lower for the proportion of the ILST that only met the USPSTF selection criteria (0.22; 95%CI: 0.07-0.71) and PLCO-exclusive subgroup had similar lung cancer risk ratios as the rest of the cohort (1.31; 95%CI: 0.82-2.08). Under the assumption that the early-stage lung cancers detected in the PLCOm2012-exclusive cohort would go undetected in the USPSTF-guided strategy – and the published modelling assumptions – the PLCOm2012-guided selection strategy is estimated to cost an additional $170 and result in an additional 0.3 QALYs over the comparator. The incremental cost-effectiveness ratio appears to be cost-effective at $561/QALY. The results were sensitive to lung cancer treatment costs and modifiable heart and lung health risks such as prevention and improved quality of life through earlier and better management of COPD; while insensitive to annual screening costs.
Conclusion
When risk-model informed screening selection is compared to the USPSTF selection guidelines, risk-model selection introduces minor costs but may prove cost-saving in the long-term, and lead to improved QALY gains. Updates to immunotherapy treatment costs are likely to counterbalance any additional costs associated with model-based risk selection.
-
+
P43 - Screening and Early Detection - Biomarkers (ID 178)
- Event: WCLC 2020
- Type: Posters
- Track: Screening and Early Detection
- Presentations: 1
- Moderators:
- Coordinates: 1/28/2021, 00:00 - 00:00, ePoster Hall
-
+
P43.02 - A Multiplex Plasma Protein Panel for Detection of Lung Cancer in Ever Smokers Not Currently Eligible for Screening (ID 2407)
00:00 - 00:00 | Author(s): Sukhinder Atkar-Khattra
- Abstract
Introduction
Lung cancer screening using low-dose computed tomography (LDCT) of the chest has been shown to reduce lung cancer mortality by 20% to 24% in heavy ever smokers. A sizeable proportion of light ever smokers do not meet screening criteria as defined by the US Preventive Services Task Force (USPSTF) for age, pack-years and quit-time. They furthermore do not meet the LDCT screening criteria set forth with the PLCOm2012 risk prediction model 6-year lung cancer risk threshold of ≥1.51%. To improve the impact of lung cancer screening programs, a simple low-cost risk-stratification method capable of identifying light ever smokers with similar lung cancer risk as heavy ever smokers is urgently needed. This study presents plasma protein assessment of informative biomarkers which may independently identify risk.
Methods
The study sample includes ever-smokers (55-80 years) with and without lung cancer who do not meet PLCOm2012 lung cancer risk model threshold of ≥1.51%/6-years or the USPSTF criteria (≥30 pack-years and smoked within the past 15 years). Plasma samples from individuals who were not eligible for the International Lung Screening Trial (ILST) and light ever smokers with Stage I lung cancer prior to surgical resection were subject to multiplex assay for Pro-surfactant protein B (ProSB) and Carcinoembryonic antigen (CEA) using the MagArray magnetic nanosensor. The results were compared to ILST participants with or without lung cancer with PLCOm2012 risk score >1.5%.
Results
The characteristics of the subjects studied are shown in Table 1.
Among the light ever smokers with a PLCOm2012 risk score <1.5%, a significantly higher level of ProSB was observed in those with lung cancer compared to the non-cancer controls (P=0.001) ProSB levels are similar between cancer subjects irrespective of the PLCOm2012 risk score but were significantly lower in the non-cancer subjects with a PLCOm2012 risk score <1.5% versus those with a PLCOm2012 risk score >1.5%. CEA was higher in the cancer versus non-cancer subjects although not statistically significant in those with a PLCOm2012 risk score <1.5% (P=0.09).
Conclusion
Using a blood biomarker such as ProSB may identify light ever smokers who do not meet current eligibility criteria for screening but may benefit from LDCT screening.
