PROMISE: a real-world clinical-genomic database to address knowledge gaps in prostate cancer

Purpose Prostate cancer is a heterogeneous disease with variable clinical outcomes. Despite numerous recent approvals of novel therapies, castration-resistant prostate cancer remains lethal. A “real-world” clinical-genomic database is urgently needed to enhance our characterization of advanced prostate cancer and further enable precision oncology. Methods The Prostate Cancer Precision Medicine Multi-Institutional Collaborative Effort (PROMISE) is a consortium whose aims are to establish a repository of de-identified clinical and genomic patient data that are linked to patient outcomes. The consortium structure includes a (1) bio-informatics committee to standardize genomic data and provide quality control, (2) biostatistics committee to independently perform statistical analyses, (3) executive committee to review and select proposals of relevant questions for the consortium to address, (4) diversity/inclusion committee to address important clinical questions pertaining to racial disparities, and (5) patient advocacy committee to understand patient perspectives to improve patients’ quality of care. Results The PROMISE consortium was formed by 16 academic institutions in early 2020 and a secure RedCap database was created. The first patient record was entered into the database in April 2020 and over 1000 records have been entered as of early 2021. Data entry is proceeding as planned with the goal to have over 2500 patient records by the end of 2021. Conclusions The PROMISE consortium provides a powerful clinical-genomic platform to interrogate and address data gaps that have arisen with increased genomic testing in the clinical management of prostate cancer. The dataset incorporates data from patient populations that are often underrepresented in clinical trials, generates new hypotheses to direct further research, and addresses important clinical questions that are otherwise difficult to investigate in prospective studies.

Establishing a genomic database for the medicinal plants in the Brazilian Pharmacopoeia

Background Brazil is exceptionally abundant in medicinal plant resources and has a rich ethnopharmacological history. Brazilian Pharmacopoeia (BP) acts as a national standard that regulates drug quality and has six published editions. Recent genomic approaches have led to a resurgence of interest in herbal drugs. The genomic data of plants has been used for pharmaceutical applications, protecting natural resources, and efficiently regulating the market. However, there are few genomic databases specifically for medicinal plants, and the establishment of a database that focuses on the herbs contained in the BP is urgently required. Methods The medicinal plant species included in each edition of the BP were analyzed to understand the evolution of the Brazilian herbal drugs. The data of 82 plants in the BP were collected and categorized into four sections: DNA barcodes, super-barcodes, genomes, and sequencing data. A typical web server architecture pattern was used to build the database and website. Furthermore, the cp-Gs of the Aloe genus in the database were analyzed as an illustration. Results A new database, the Brazilian Pharmacopoeia Genomic Database (BPGD) was constructed and is now publicly accessible. A BLAST server for species identification and sequence searching with the internal transcribed spacer 2 (ITS2), the intergenic region (psbA-trnH), and the chloroplast genome (cp-G) of Brazilian medicinal plants was also embedded in the BPGD. The database has 753 ITS2 of 76 species, 553 psbA-trnH and 190 genomes (whole genome and chloroplast genome) of 57 species. In addition, it contains 37 genome sequence data sets of 24 species and 616 transcriptome sequence data sets of 34 species and also includes 187 cp-Gs representing 57 medicinal species in the BP. Analyses of the six cp-Gs of three Aloe species identified the variable regions in the cp-Gs. These can be used to identify species and understand the intraspecific relationships. Conclusions This study presents the first genomic database of medicinal plants listed in the latest BP. It serves as an efficient platform to obtain and analyze genomic data, accelerate studies regarding Brazilian medicinal plants and facilitate the rational development on their market regulation.

Establishing a Genomic Database for the Traditional Medicinal Plants in the Brazilian Pharmacopoeia

BackgroundBrazil is exceptionally abundant in traditional medicinal plant resources and has a rich ethnopharmacological history. Brazilian Pharmacopoeia (BP) acts as a national standard regulating drugs’ quality and has six published editions. Recent genomic approaches have led to a resurgence of interest in herbal drugs. Plants’ genomic data have been used for pharmaceutical applications, protecting natural resources, and efficiently regulating the market. However, there are few genomic databases specifically on medicinal plants, and establishing one focusing on the herbs of BP is urgently needed. MethodsThe BP editions’ medicinal plant species were analyzed to understand the evolution of the herbal drugs in Brazil. A new database, BPGD (Brazilian Pharmacopoeia Genomic Database), was constructed based on a typical web server architecture. A BLAST server for species identification and sequence searching with internal transcribed spacer 2 (ITS2), intergenic region (psbA-trnH), and chloroplast genome (cp-G) of Brazilian traditional medicinal plants was also embedded in BPGD. Data of 82 plants in BP were collected and categorized into four parts: DNA barcodes, super-barcodes, genomes, and sequencing data. Further, the cp-Gs of Aloe genus in the database were analyzed as an illustration.ResultsBPGD (V1.0) has been tested and opened for public users. The database provides a comprehensive set of data, including the description and identification criteria of medicinal plants, and allows sequence-based search using BLAST. The database has 753 ITS2 of 76 species, 553 psbA-trnH and 190 genomes (whole genome and chloroplast genome) of 57 species, and 37 genome sequence data sets of 24 species and 616 transcriptome sequence data sets of 34 species. The data includes 187 cp-Gs representing 57 medicinal species in BP. Analysis of the six cp-Gs of three Aloe species identified the variable regions in cp-Gs, which could be used to identify species and understand the intraspecific relationship. ConclusionsThis study presents the first genomic database for traditional medicinal plants listed in the latest BP. It serves as an efficient platform to obtain genomic data, specifically on medical plants listed in the BP (http://bpgenome.com/).

Impact of STK11 mutation on first-line immune checkpoint inhibitor (ICI) outcomes in a real-world KRAS G12C mutant lung adenocarcinoma cohort.

9106 Background: The introduction of KRAS G12C inhibitors into clinical trials has demonstrated promise and may provide a new therapeutic option for patients (pts) harboring KRAS G12C mutations. Recent data has also indicated that immune checkpoint inhibitors (ICI) have shown benefit in KRAS G12C mutant lung adenocarcinoma (LUAD); however, data on the impact of co-occurring STK11 mutations on outcomes are conflicting. We utilized the Guardant INFORM real-world clinical-genomic database to assess the impact of co-occurring STK11 mutations on outcomes in pts with KRAS G12C mutant LUAD treated with a first-line ICI containing regimen. Methods: This retrospective matched cohort observational study was conducted in a nationally representative clinical-genomic database covering over 137,000 pts with comprehensive ctDNA results and associated clinical information. Adult pts with metastatic LUAD who received ≥ 1 dose of first-line anti-PD1/PD-L1 ± chemotherapy and had at least 90 days follow-up after first KRAS G12C detection were included. A cohort of pts without KRAS G12C, including KRAS wildtype pts and pts with other KRAS mutations, were matched 3:1 for age, gender, year of index and baseline comorbidity. Time to next treatment (TTNT), time to discontinuation (TTD), real-world overall survival (rwOS) were compared with vs. without STK11 mutations for both cohorts using cox proportional-hazards model. Results: Among 330 pts in the KRAS G12C cohort, 21% (n = 70) had an STK11 mutation. Among the matched cohort (n = 938), 754 pts were KRAS wildtype, of whom 6% (n = 49) had STK11 mutations. Within the KRAS G12C cohort, pts with STK11 mutations had statistically significant shorter TTNT (hazard ratio [HR] 2.7, 95% confidence internal [CI] 1.8-4.0, p < 0.0001), TTD (HR 1.4, 95% CI 1.0-2.0, p < 0.04) and rwOS (HR 3.2, 95% CI 2.0-5.1, p < 0.0001) than pts without STK11 mutations. Within the matched KRAS wildtype cohort, the differences in TTD (HR 1.4, 95% CI = 1.0-2.0, p = 0.08) and rwOS (HR 1.4, 95% CI = 0.8-2.4, p = 0.3) in patients with vs. without STK11 mutation did not reach statistical significance (Table). Conclusions: This study provides real-world evidence that STK11 co-mutations are associated with worse outcomes among pts with KRAS G12C mutant LUAD treated with first-line ICI. These inferior outcomes indicate a high unmet medical need among LUAD pts harboring co-occurring KRAS G12C and STK11 mutations and demonstrate the need for effective targeted and/or combination therapies in this patient population.[Table: see text]