Start Your Search
JCSE01 - Perspectives for Lung Cancer Early Detection (ID 779)
- Event: WCLC 2018
- Type: Joint IASLC/CSCO/CAALC Session
- Track: Screening and Early Detection
- Presentations: 1
- Coordinates: 9/23/2018, 07:30 - 11:15, Room 202 BD
JCSE01.06 - Incorporating Artificial Intelligence for Early Detection of Lung Cancer (ID 11399)
09:20 - 09:40 | Presenting Author(s): Jie Hu
Abstract not provided
Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.
P1.11 - Screening and Early Detection (Not CME Accredited Session) (ID 943)
- Event: WCLC 2018
- Type: Poster Viewing in the Exhibit Hall
- Presentations: 1
- Coordinates: 9/24/2018, 16:45 - 18:00, Exhibit Hall
P1.11-01 - The NELSON Triage Algorithm Applied to a Chinese Biopsied Population: A Pilot Study (ID 12115)
16:45 - 18:00 | Author(s): Jie Hu
The Nederlands-Leuvens Longkanker Screenings Onderzoek (NELSON) lung cancer screening study sequentially tests nodules’ volume, growth and doubling time with an increasing screening interval length. For now, classification performances of NELSON are known only for a European population with specific eligibility criteria.
We tested the NELSON trial triage algorithm on a cohort of biopsied Chinese patients.a9ded1e5ce5d75814730bb4caaf49419 Method
Our study utilized a data subset from the NCT02693496 clinical trial. The data consisted of onsite prospective evaluations of 85 Chinese patients who underwent CTs within an average time interval of 259 days (Min=63; Max=1092). NCT02693496 readers applied the Chinese consensus low-dose CT management guidelines with referral to biopsy, when required.
The eligibility criteria were: All genders; age: 18 to 90 years old; chest lesion <3cm. Smoking status was not considered.
In our subset of 85 patients, 15 nodules from 15 patients were biopsied: 10 were confirmed malignant; of these, 3 were solid nodules (SN) and 7 were sub-solid nodules (SSN). Five biopsied nodules were confirmed benign. Of the whole cohort, 11.8% (10/85), were declared positive patients. The 75/85 others (88.2%) were declared negative by radiologists and/or pathologists.
Using the Lesion Management Solution (LMS) platform (Median Technologies), we retrospectively re-processed the subset of images to analyze NELSON sensitivity at detecting malignant lung nodules. We used R CRAN software for statistics and Chi-Squared test for non-parametric comparison of two sample proportions.4c3880bb027f159e801041b1021e88e8 Result
We found 12.9% of patients (11/85) displayed no findings. There were 155 detected findings in the remaining 74 patients, which were documented as: 5.2% (8/155) benign as NODCAT I; 9.0% (14/155) pleural; 27.7% (43/155) SSN and 58.1% (90/155) SN. According to NELSON triage, five Patients were declared positive.
In the biopsy-confirmed nodules group (10 patients), four were detected by NELSON, one at baseline.
All patients displaying SNs were detected (3/3) whereas only one (1/6) patient displaying SSNs was detected. NELSON was better at detecting SNs (p=0.036).
One patient with confirmed negative biopsy patient (1/5) was declared positive by NELSON with one SN and one SSN.
Additionally, one SN in patient without biopsy was declared positive at baseline by NELSON triage algorithm.8eea62084ca7e541d918e823422bd82e Conclusion
Prevalence of different nodule types in this Chinese population was different from the observations of the original European NELSON trial. The NELSON triage algorithm correctly classified patients with malignant SNs but misclassified most patients displaying SSNs. Further studies are needed to better evaluate SSNs by NELSON.6f8b794f3246b0c1e1780bb4d4d5dc53