Abstract 1581P
Background
Navya-AI is a clinically validated digital intervention that outputs patient-specific, evidence-based information; with 97% concordance between Navya-AI output and expert recommendations, and 80% adoption of recommendations by patients. Given high financial toxicity of suboptimal maintenance treatments in OC, use of AI to encourage optimal treatment has potential benefit. This study assesses the impact of Navya-AI in improving compliance to guidelines for genomic testing for women with OC in India; especially since generics for genomic testing and PARP inhibitors are cost effectively accessible in India.
Methods
Since March 2022, genomic testing for BRCA1/2 and HRD has been a recommended guideline for stage III & IV OC by NCCN, ESMO, and NCG India. From March 2022 to March 2024, all women with OC who received a Navya-AI enabled review of their treatment plan were prospectively analyzed for concordance on genomic testing and precision care. Intervention with Navya AI was used to close any identified care gaps.
Results
Of 260 OC patients who received a Navya-AI review, 80% [209/260] met guideline criteria for genomic testing. Stage I/II [47], and rare histologies [4] were excluded. Women were diverse with respect to age (years) [0-35: 6%, 35-50: 30%, 51-65: 49%, >65: 15%]; stage [III 42%, IV: 58%]; family history of cancer: 24%. Genomic testing and precision care was planned for only 30% [63/209] of the patients. In the remaining 70% [146/209], digital intervention by Navya-AI including information on the risks/benefits of genomic testing, potential use of PARP inhibitors, importance of cascade genetics, and patient navigation to their treating oncologists enabled guidelines compliance.
Conclusions
Guideline compliant care is a useful metric in tracking quality of care in gynecologic cancers, where wide care disparities exist. Positive findings on genomic testing in OC has a significant impact on progression-free survival and decreasing cancer burden. More than 60% of patients presenting to a nationally used digital health expert opinion service in India would have missed such critical testing. Through AI driven technologies, this disparity can be identified and closed, globally.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
Has not received any funding.
Disclosure
All authors have declared no conflicts of interest.
Resources from the same session
1607P - Association of the lipid biomarker, PCPro, and clinical outcomes in the ENZAMET trial (ANZUP 1304)
Presenter: Lisa Horvath
Session: Poster session 10
1608P - Prostate cancer working group 4 (PCWG4) preliminary criteria using serial PSMA PET/CT for response evaluation: Analysis from the PRINCE trial
Presenter: Michael Hofman
Session: Poster session 10
1609P - PSMA-PET and PROMISE re-define stage and risk in patients with prostate cancer
Presenter: Wolfgang Fendler
Session: Poster session 10
1610P - Circulating tumour cell (CTC) enumeration and overall survival (OS) in men with metastatic castration-resistant prostate cancer (mCRPC) treated with xaluritamig
Presenter: Andrew Armstrong
Session: Poster session 10
1611P - Haematologic impact of [177Lu]Lu-PSMA-617 versus ARPI change in patients with metastatic castration-resistant prostate cancer in PSMAfore
Presenter: Kim Nguyen Chi
Session: Poster session 10
1612P - Impact of FANCA, ATM, CDK12 alterations on survival in metastatic castration-resistant prostate cancer (mCRPC)
Presenter: David Lorente
Session: Poster session 10
1613P - Clinically advanced prostate cancer (CAPC) featuring BRCA2 loss: A comprehensive genomic profiling (CGP) study
Presenter: Chiara Mercinelli
Session: Poster session 10
1614P - PSA responses and PSMA scan changes after immunotherapy for biochemically recurrent prostate cancer (BCR) without androgen deprivation therapy (ADT)
Presenter: Ravi Madan
Session: Poster session 10
1615P - A new prognostic model of overall survival (OS) in patients (pts) with metastatic hormone sensitive prostate cancer (mHSPC)
Presenter: Susan Halabi
Session: Poster session 10