Abigail Fong1, Rebecca Nelson1, Madhurima Datta2, Alex Exuzies2, Yosef Chodakiewitz2, Daniel Durand2, Jae Kim1, Stacy Gray1, Dan Raz1
1 City of Hope, Duarte, CA, United States, 2 Prenuvo, Los Angeles, CA, United States Purpose
Patients are increasingly taking control of their health care and diagnostics. Although not guideline recommended, whole-body MRI (WB-MRI) is increasingly utilized for preventative health assessment, particularly among individuals concerned about their cancer risk. This study assesses the prevalence and significance of abnormal pulmonary and mediastinal findings in WB-MRI screening.
Methods
We conducted a retrospective analysis of 118,019 WB-MRI scans performed at Prenuvo Clinics in the U.S. and Canada between January 2020 to June 2025. The scans were done using standardized non-contrast multiparametric protocols on 1.5T Siemens (25%) or Philips (75%) scanners optimized to a 45-60 minute session. All protocols included STIR, T1, Dixon, DWI, T2; WB and WT added Brain FLAIR and TOF MRA. Exams were interpreted by specialized radiologists trained in WB-MRI and reported using ONCO-RADS structured reporting for WB-MRI, with findings categorized by organ system and ONCO-RADS category. Demographics, smoking history, family history, and medical history were obtained from patient-reported medical intake forms. Results were analyzed using descriptive statistics and include all patients with lung nodules who were free of known active lung cancer at the time of their scan.
Results
3,547 lung and mediastinal nodules were reported on 3,524 scans from 3,388 individuals. 3,434 (96.8%) were lung lesions, 77 (2.2%) mediastinal lesions, 36 (1.0%) indeterminant. Median age was 54 years (IQR: 43-64 years). 57% were female. 46.4% were White. Median BMI was 24.3 (IQR: 21.6-27.3).
Median lung lesion size was 7mm (IQR: 5-10mm); of these 1,066 (31%) were <6mm, 1,014 (29.5%) were 6- 8mm, and 1,201 (35%) were >8mm; 153 (4.5%) of scans did not include size. In 232 (6.8%) scans, lesions were evaluated ONCO-RADs 4/ 5 (potentially/highly oncologically suspicious). Median lesion size in ONCO RADs 4-5 was 17mm (IQR: 12-27mm), mean age was 58.1 years (±13). 37.3% had a tobacco use history. In patients with pulmonary findings, 31.1% had a tobacco use history (median pack-years: 5, IQR: 1.6-15).
In 798 (24.1%) lung lesion patients with <20 pack-year tobacco use history, median nodule size was 7mm (IQR: 5-10mm); for 178 (5.4%) with ≥20 pack-year history, median nodule size was 7mm (IQR: 5-12mm) (p=0.28).
In 76 patients with mediastinal masses, mean age was 50.1 (±13.2), 59.2% were female. Of 77 masses, 41 (53.2%) were anterior mediastinal masses, 20 (26.0%) were benign appearing cysts and 53 (68.8%) were concerning for malignancy.
Conclusions
This study is the largest study of pulmonary and mediastinal findings on WB-MRI. Relatively high rates of potentially cancerous intra-thoracic findings can be identified on WB-MRI, including in patients not eligible for lung cancer screening. Further research is needed to determine the utility of WB-MRI in reducing cancer mortality.
Bibliographic: Owens C, et al. The lung cancers: staging and response, CT, (18)F-FDG PET/CT, MRI, DWI: review and new perspectives. BR J Radiol. 2023; 96 (1148). Biederer J, et al. Screening for lung cancer: Does MRI have a role? Eur J Radiol. 2017 Jan; 86:353-360.
Disclosure: Abigail Fong: None declared, Rebecca Nelson: None declared, Madhurima Datta: Prenuvo, Advisory Board: No, Consultant: No, Owner/Co-Owner Founder/Co-Founder: No, Ownership Interest (stocks, stock options): No, Employee: Yes, Ownership Interest (patent or other intellectual property or other ownership interest excluding diversified mutual funds): No, Royalties: No, Speaker/Honoraria: No, Proctor: No, Research Grant: No, Alex Exuzies: Prenuvo, Advisory Board: No, Consultant: No, Owner/Co Owner Founder/Co-Founder: No, Ownership Interest (stocks, stock options): No, Employee: Yes, Ownership Interest (patent or other intellectual property or other ownership interest excluding diversified mutual funds): No, Royalties: No, Speaker/Honoraria: No, Proctor: No, Research Grant: No, Yosef Chodakiewitz: Prenuvo, Advisory Board: No, Consultant: No, Owner/Co-Owner Founder/Co Founder: No, Ownership Interest (stocks, stock options): No, Employee: Yes, Ownership Interest (patent or other intellectual property or other ownership interest excluding diversified mutual funds): No, Royalties: No, Speaker/Honoraria: No, Proctor: No, Research Grant: No, Daniel Durand: Prenuvo, Advisory Board: No, Consultant: No, Owner/Co-Owner Founder/Co-Founder: No, Ownership Interest (stocks, stock options): No, Employee: Yes, Ownership Interest (patent or other intellectual property or other ownership interest excluding diversified mutual funds): No, Royalties: No, Speaker/Honoraria: No, Proctor: No, Research Grant: No, Jae Kim: None declared, Stacy Gray: None declared, Dan Raz: None declared
