The application of radiomics to the setting of lung cancer screening has revealed peritumoral and intratumoral features of early-stage lung cancer tumors that are associated with worse overall survival (OS), according to a study published in Scientific Reports.1

Although annual lung cancer screening of high-risk individuals using low-dose helical computed tomography (LDCT) imaging has been associated with a substantial reduction in lung-cancer related death,2 results of some studies have suggested that it may also be associated with overdiagnosis and overtreatment in individuals with more indolent forms of lung cancer. Hence, there is a need for imaging-related biomarkers that can help distinguish between aggressive and indolent lung cancer lesions.

Radiomics is a noninvasive method that involves the identification and extraction of specific radiographic features from standard-of-care radiographic images. These radiographic features, corresponding to regions within tumor (ie, intratumoral) or surrounding the tumor parenchyma (eg, peritumoral), can potentially be used to facilitate disease diagnosis, risk assessment, and prediction of response to treatment.

This analysis utilized deidentified data and LDCT images for 234 individuals with screen-detected lung cancer included in the National Cancer Institute (NCI) Cancer Data Access System who had participated in the National Lung Cancer Screening Trial. All individuals had a baseline negative screen, but were subsequently diagnosed with lung cancer following the next 1 or 2 annual LDCT scans. Individuals in this dataset were randomly assigned to a training cohort (161 individuals), and a test cohort (73 individuals) for validation. In addition, an external validation cohort of 62 patients with resected lung adenocarcinoma not detected through LDCT screening but for whom CT imaging was performed within 2 months of surgery was also included in the study.


Continue Reading

A key finding of this study was that 2 radiomic features — an intratumoral feature associated with intensity statistical features and a peritumoral feature associated with texture relating to the rate of intensity change within an image — were useful in stratifying patients with early-stage lung cancer into low-, intermediate-, and high-risk groups.

Specifically, for patients with early-stage disease, the hazard ratio for OS for the high-risk group was 9.91 compared with the low-risk group (P <.0001). In addition, 5-year OS was determined to be 0% and 78% for those at high- and low-risk, respectively, according to the radiomic model. These results were validated for patients with early-stage lung cancer.

Furthermore, using this radiomics model, an analysis of presurgical CT scans and tumor genomic profiling performed on another dataset of 103 patients with resected adenocarcinoma from a different study suggested that the intratumoral statistical intensity radiomic feature was associated with expression of FOXF2, a modulator of gene transcription.

In their concluding remarks, the study authors commented that “the results from our analyses produced a parsimonious radiomic model that identified a vulnerable subset of screen-detected lung cancers that are associated with poor outcome.”

They further added that “these findings could support more aggressive treatment and follow-up for such high-risk patients. Nonetheless, additional research will be needed to inform the potential translational implications of these findings, to fully elucidate the biology these high-risk screen-detected tumors, to assess whether these findings are consistent across screening trials and cohorts, and how best to personalize cancer management in these vulnerable patients.”

References

  1. Perez-Morales J, Tunali I, Stringfield O, et al. Peritumoral and intratumoral radiomic features predict survival outcomes among patients diagnosed in lung cancer screening [published online June 29, 2020]. Sci Rep. doi: 10.1038/s41598-020-67378-8
  2. Alberle DR, Adams AM, Berg CD, et al. Reduced lung-cancer mortality with low-dose computed tomographic screening. N Engl J Med. 2011;365;395-409.