Virtual Library
Start Your Search
C. Parsy-Kowalska
Author of
-
+
P2.13 - Radiology/Staging/Screening (ID 714)
- Event: WCLC 2017
- Type: Poster Session with Presenters Present
- Track: Radiology/Staging/Screening
- Presentations: 1
- Moderators:
- Coordinates: 10/17/2017, 09:30 - 16:00, Exhibit Hall (Hall B + C)
-
+
P2.13-013 - Determination of the Detection Lead Time for Autoantibody Biomarkers in Early Stage Lung Cancer Using the UKCTOCS Cohort (ID 9999)
09:30 - 09:30 | Author(s): C. Parsy-Kowalska
- Abstract
Background:
Tumor associated (TA) autoantibodies are present during early stage lung cancer and have been detected up to five years before diagnosis. However the detection lead time provided by their measurement has never been accurately determined due to a lack of suitable patient samples. The United Kingdom Collaborative Trial of Ovarian Cancer Screening (UKCTOCS) recruited 202,638 postmenopausal women. Annual blood samples were collected for over 10 years during which time a number of lung cancer cases were diagnosed. The primary aim of this study was to determine the detection lead time of tumor associated autoantibody assays for a subset of the UKCTOCS cohort.
Method:
A set of 142 primary lung cancer cases (NSCLC 83%, SCLC 12%) with 7 serial samples over a pre-diagnosis period of 10 years were randomly split into Training (n=100) and Validation (n=42) cohorts and matched to healthy controls by age-at-diagnosis, age-at-first sample and smoking history. TA autoantibody profiles were produced for each patient by measuring autoantibody levels against a panel of tumour associated antigens ( p53, SOX2, CAGE, NY-ESO-1, GBU4-5, MAGE A4, HuD, CK8, CK20, LMYC, SSX1, p53-95, p16 and p62) using ELISA. The profiles for each patient were compared to those for the preceding time point using a multivariate distance measure to determine if a statistically significant positive change had occurred. Comparison against a population based cut-off for the earliest time point sample for each patient was used to determine initial positivity. An optimised algorithm was developed on the Training cohort and then applied to the Validation cohort.
Result:
There were 49 positive patients in the training cohort: 11 at the earliest time point and 38 during serial sampling. For the Validation cohort there were 14: 3 at the earliest and 11 during serial sampling. The median detection lead time for the Training cohort was 4.0 years (0.3 to 9.4 range) and for the Validation cohort 4.3 years (0.1 to 9.0 range) before clinical diagnosis. The median was 4.1 years (0.1 to 9.4 range) for the entire cohort.
Conclusion:
This is the first time statistically sound estimates of detection lead time have been reported for tumor assoicated autoantibody tests run on such a large cohort of pre-diagnostic serial samples. These cancer biomarkers can be detected on average 4 years before diagnosis. Monitoring autoantibody profiles could be hugely beneficial by enabling earlier detection and stratification of screening populations for cancer. This could lower mortality rates and reduce healthcare costs.
