Artificial INtelligence to Support Informed DEcision-making (INSIDE) for Improved Literature Analysis in Oncology.

BACKGROUND: Defining optimal therapeutic sequencing strategies in prostate cancer (PC) is challenging and may be assisted by artificial intelligence (AI)-based tools for an analysis of the medical literature. OBJECTIVE: To demonstrate that INSIDE PC can help clinicians query the literature on therapeutic sequencing in PC and to develop previously unestablished practices for evaluating the outputs of AI-based support platforms. DESIGN, SETTING, AND PARTICIPANTS: INSIDE PC was developed by customizing PubMed Bidirectional Encoder Representations from Transformers. Publications were ranked and aggregated for relevance using data visualization and analytics. Publications returned by INSIDE PC and PubMed were given normalized discounted cumulative gain (nDCG) scores by PC experts reflecting ranking and relevance. INTERVENTION: INSIDE PC for AI-based semantic literature analysis. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: INSIDE PC was evaluated for relevance and accuracy for three test questions on the efficacy of therapeutic sequencing of systemic therapies in PC. RESULTS AND LIMITATIONS: In this initial evaluation, INSIDE PC outperformed PubMed for question 1 (novel hormonal therapy [NHT] followed by NHT) for the top five, ten, and 20 publications (nDCG score, +43, +33, and +30 percentage points [pps], respectively). For question 2 (NHT followed by poly [adenosine diphosphate ribose] polymerase inhibitors [PARPi]), INSIDE PC and PubMed performed similarly. For question 3 (NHT or PARPi followed by 177Lu-prostate-specific membrane antigen-617), INSIDE PC outperformed PubMed for the top five, ten, and 20 publications (+16, +4, and +5 pps, respectively). CONCLUSIONS: We applied INSIDE PC to develop standards for evaluating the performance of AI-based tools for literature extraction. INSIDE PC performed competitively with PubMed and can assist clinicians with therapeutic sequencing in PC. PATIENT SUMMARY: The medical literature is often very difficult for doctors and patients to search. In this report, we describe INSIDE PC-an artificial intelligence (AI) system created to help search articles published in medical journals and determine the best order of treatments for advanced prostate cancer in a much better time frame. We found that INSIDE PC works as well as another search tool, PubMed, a widely used resource for searching and retrieving articles published in medical journals. Our work with INSIDE PC shows new ways in which AI can be used to search published articles in medical journals and how these systems might be evaluated to support shared decision-making.

Application of artificial intelligence to overcome clinical information overload in urological cancer.

OBJECTIVE: To describe the use of artificial intelligence (AI) in medical literature and trial data extraction, and its applications in uro-oncology. This bridging review, which consolidates information from the diverse applications of AI, highlights how AI users can investigate more sophisticated queries than with traditional methods, leading to synthesis of raw data and complex outputs into more actionable and personalised results, particularly in the field of uro-oncology. METHODS: Literature and clinical trial searches were performed in PubMed, Dimensions, Embase and Google (1999-2020). The searches focussed on the use of AI and its various forms to facilitate literature searches, clinical guidelines development, and clinical trial data extraction in uro-oncology. To illustrate how AI can be applied to address questions about optimising therapeutic decision making and individualising treatment regimens, the Dimensions-linked information platform was searched for 'prostate cancer' keywords (76 publications were identified; 48 were included). RESULTS: AI offers the promise of transforming raw data and complex outputs into actionable insights. Literature and clinical trial searches can be automated, enabling clinicians to develop and analyse publications expeditiously on complex issues such as therapeutic sequencing and to obtain updates on documents that evolve at the pace and scope of the landscape. An AI-based platform inclusive of 12 trial databases and >100 scientific literature sources enabled the creation of an interactive visualisation. CONCLUSION: As the literature and clinical trial landscape continues to grow in complexity and with increasing speed, the ability to pull the right information at the right time from different search engines and resources, while excluding social media bias, becomes more challenging. This review demonstrates that by applying natural language processing and machine learning algorithms, validated and optimised AI leads to a speedier, more personalised, efficient, and focussed search compared with traditional methods.

Treatment patterns and survival in metastatic castration-sensitive prostate cancer in the US Veterans Health Administration.

BACKGROUND: Limited real-world data exist on treatment patterns and outcomes in patients with metastatic castration-sensitive prostate cancer (mCSPC). METHODS: A retrospective cohort study was conducted, using the Veterans Health Administration claims database (April 2013-March 2018). Among 369,734 prostate cancer patients, we selected all men who developed metastases within 90 days before or after medical/surgical castration and who received androgen deprivation therapy (ADT). Patients were categorized into four cohorts: ADT-only (± <90-day nonsteroidal anti-androgen [NSAA] use), ADT + NSAA, ADT + docetaxel, and ADT + abiraterone. Main outcomes were treatment patterns, time-to-progression to metastatic castration-resistant disease, and overall survival. Multivariable analysis and sensitivity analysis were conducted. RESULTS: Of 1395 patients, 874 (63%) received ADT-only, 338 (24%) received ADT + NSAA, 108 (8%) received ADT + docetaxel, and 75 (5%) received ADT + abiraterone. Proportions on ADT-only and ADT + NSAA declined (from 66% to 60% and from 31% to 17%, respectively) over the study period, while proportions prescribed ADT + docetaxel or abiraterone increased from 3% to 9% and from 1% to 15%, respectively. Patients treated with ADT + NSAA had similar risks of castration-resistant disease (hazard ratio [HR] 1.05; 95% confidence interval [CI]: 0.87, 1.26) and overall mortality (HR 1.22; 95% CI: 0.97, 1.54) as ADT-only. CONCLUSIONS: Most patients with mCSPC initiating ADT received ADT-only or ADT + NSAA, despite the emergence of docetaxel and novel hormonal therapies. Even in the most recent period (2017 to early 2018), only 24% of men received intensified therapy with agents known to prolong survival versus ADT-only. These data in real-world clinical practice suggest substantial room for improved outcomes in patients with mCSPC.

Real-world Experience With Sunitinib Treatment in Patients With Metastatic Renal Cell Carcinoma: Clinical Outcome According to Risk Score.

BACKGROUND: ADONIS is an ongoing observational study in 9 European countries, designed to evaluate treatment patterns/outcomes in patients with metastatic renal cell carcinoma (mRCC) treated with first-line sunitinib and/or second-line axitinib post sunitinib. We present an evaluation of sunitinib efficacy by risk group, in the real-world setting examined in ADONIS. PATIENTS AND METHODS: Patients were enrolled at the start of first-line sunitinib treatment or second-line axitinib post sunitinib treatment. Evaluation of sunitinib efficacy was assessed by International Metastatic Renal Cell Carcinoma Database Consortium (IMDC) and Memorial Sloan Kettering Cancer Center risk criteria. RESULTS: For all patients in this analysis (N = 467), the median progression-free survival was 23.8 months (95% confidence interval [CI], 16.5-28.5 months), 11.8 months (95% CI, 8.1-17.4 months), and 4.6 months (95% CI, 2.5-7.7 months) for IMDC favorable-, intermediate-, and poor-risk groups, respectively. The median overall survival was 97.1 months (95% CI, 46.3 months-not evaluable [NE]), 33.5 months (95% CI, 20.5-46.6 months), and 10.0 months (95% CI, 4.5-19.8 months) for the respective risk groups. Data on individual risk factors were available for a subgroup of patients, allowing analysis by intermediate risk by 1 versus 2 risk factors. When including this subgroup (n = 120), the median overall survival for IMDC favorable-, intermediate-1, and intermediate-2 risk factors was 21.6 months (95% CI, 16.3 months-NE), 20.5 months (15.5 months-NE), and 15.1 months (4.1 months-NE), respectively. CONCLUSIONS: For patients overall and by risk-group stratification, survival estimates were aligned with previously published data. In patients with intermediate-1 risk, overall survival was very similar to patients with favorable risk. However, further exploration of outcome data from different sources is needed to confirm these observations.

Safety and effectiveness of classical and alternative sunitinib dosing schedules for metastatic renal cell carcinoma: a meta-analysis.

The optimal dosing schedule to maintain the effectiveness of sunitinib for metastatic renal cell carcinoma - while reducing toxicity - remains an important clinical question. A meta-analysis of randomized trials and observational studies assessed the relative treatment effects of 4/2, 2/1 and transitional-2/1 schedules on outcomes and adverse events using Bayesian network meta-analysis methods. Treatment with 2/1 reduced the risk of disease progression or death by 25% and had lower odds of hand-and-foot syndrome compared with the 4/2. A numerical but not 'statistical' benefit in progression-free survival was observed with the transitional-2/1 compared with 4/2. Alternative schedules with the 2/1 and transitional-2/1 may be more clinically beneficial in metastatic renal cell carcinoma than the 4/2 schedule.

Sunitinib in Patients With Metastatic Renal Cell Carcinoma: Clinical Outcome According to International Metastatic Renal Cell Carcinoma Database Consortium Risk Group.

BACKGROUND: Sunitinib malate, a targeted tyrosine kinase inhibitor, is standard of care for metastatic renal cell carcinoma (mRCC) and serves as the active comparator in several ongoing mRCC clinical trials. In this analysis we report benchmarks for clinical outcomes on the basis of International Metastatic Renal Cell Carcinoma Database Consortium (IMDC) risk groups for patients treated with sunitinib for mRCC in a first-line setting. MATERIALS AND METHODS: A retrospective analysis was performed on data from sunitinib-treated patients (n = 375) in the pivotal phase III trial of sunitinib versus interferon-α as first-line treatment for mRCC. Objective response rates (ORRs) were determined from independently reviewed radiologic assessments. The Kaplan-Meier method was used to estimate median progression-free survival (PFS) and median overall survival (OS) according to patient risk group. RESULTS: Median PFS (95% confidence interval [CI]) was 14.1 (13.4-17.1), 10.7 (10.5-12.5), 2.4 (1.1-4.7), and 10.6 (8.1-10.9) months in sunitinib-treated patients in the IMDC favorable (n = 134), intermediate (n = 205), poor (n = 34), and intermediate + poor (n = 239) risk groups, respectively. Median OS (95% CI) was 23.0 (19.8-27.8), 5.1 (4.3-9.9), and 20.3 (16.8-23.0) months in sunitinib-treated patients in IMDC intermediate, poor, and intermediate + poor risk groups, respectively, and was not reached in the favorable risk group (>50% of patients were alive at data cutoff). ORRs (95% CI) was 53.0% (44.2%-61.7%), 33.7% (27.2%-40.6%), 11.8% (3.3%-27.5%), and 30.5% (24.8%-36.8%) in sunitinib-treated patients in IMDC favorable, intermediate, poor, and intermediate + poor risk groups, respectively. CONCLUSION: Results of this retrospective analysis show differences in patient outcomes for PFS, OS, and ORR on the basis of IMDC prognostic risk group assignment for patients with mRCC.

Adjuvant Sunitinib for High-risk Renal Cell Carcinoma After Nephrectomy: Subgroup Analyses and Updated Overall Survival Results.

BACKGROUND: Adjuvant sunitinib significantly improved disease-free survival (DFS) versus placebo in patients with locoregional renal cell carcinoma (RCC) at high risk of recurrence after nephrectomy (hazard ratio [HR] 0.76, 95% confidence interval [CI] 0.59-0.98; p=0.03). OBJECTIVE: To report the relationship between baseline factors and DFS, pattern of recurrence, and updated overall survival (OS). DESIGN, SETTING, AND PARTICIPANTS: Data for 615 patients randomized to sunitinib (n=309) or placebo (n=306) in the S-TRAC trial. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Subgroup DFS analyses by baseline risk factors were conducted using a Cox proportional hazards model. Baseline risk factors included: modified University of California Los Angeles integrated staging system criteria, age, gender, Eastern Cooperative Oncology Group performance status (ECOG PS), weight, neutrophil-to-lymphocyte ratio (NLR), and Fuhrman grade. RESULTS AND LIMITATIONS: Of 615 patients, 97 and 122 in the sunitinib and placebo arms developed metastatic disease, with the most common sites of distant recurrence being lung (40 and 49), lymph node (21 and 26), and liver (11 and 14), respectively. A benefit of adjuvant sunitinib over placebo was observed across subgroups, including: higher risk (T3, no or undetermined nodal involvement, Fuhrman grade ≥2, ECOG PS ≥1, T4 and/or nodal involvement; hazard ratio [HR] 0.74, 95% confidence interval [CI] 0.55-0.99; p=0.04), NLR ≤3 (HR 0.72, 95% CI 0.54-0.95; p=0.02), and Fuhrman grade 3/4 (HR 0.73, 95% CI 0.55-0.98; p=0.04). All subgroup analyses were exploratory, and no adjustments for multiplicity were made. Median OS was not reached in either arm (HR 0.92, 95% CI 0.66-1.28; p=0.6); 67 and 74 patients died in the sunitinib and placebo arms, respectively. CONCLUSIONS: A benefit of adjuvant sunitinib over placebo was observed across subgroups. The results are consistent with the primary analysis, which showed a benefit for adjuvant sunitinib in patients at high risk of recurrent RCC after nephrectomy. PATIENT SUMMARY: Most subgroups of patients at high risk of recurrent renal cell carcinoma after nephrectomy experienced a clinical benefit with adjuvant sunitinib. TRIAL REGISTRATION: ClinicalTrials.gov NCT00375674.

Small molecule obatoclax (GX15-070) antagonizes MCL-1 and overcomes MCL-1-mediated resistance to apoptosis.

Elevated expression of members of the BCL-2 pro-survival family of proteins can confer resistance to apoptosis in cancer cells. Small molecule obatoclax (GX15-070), which is predicted to occupy a hydrophobic pocket within the BH3 binding groove of BCL-2, antagonizes these members and induces apoptosis, dependent on BAX and BAK. Reconstitution in yeast confirmed that obatoclax acts on the pathway and overcomes BCL-2-, BCL-XL-, BCL-w-, and MCL-1-mediated resistance to BAX or BAK. The compound potently interfered with the direct interaction between MCL-1 and BAK in intact mitochondrial outer membrane and inhibited the association between MCL-1 and BAK in intact cells. MCL-1 has been shown to confer resistance to the BCL-2/BCL-XL/BCL-w-selective antagonist ABT-737 and to the proteasome inhibitor bortezomib. In both cases, this resistance was overcome by obatoclax. These findings support a rational clinical development opportunity for the compound in cancer indications or treatments where MCL-1 contributes to resistance to cell killing.

Annexin-I expression modulates drug resistance in tumor cells.

The use of anti-cancer chemotherapy often leads to the rise of multidrug-resistant (MDR) tumors. We have previously reported the overexpression of a 40kDa protein (P-40) in several MDR tumor cell lines. In this report we describe the cloning of a 1.4kb cDNA with an open reading frame of 344 amino acids that encodes the P-40 protein. Analysis of the P-40 amino acid sequence showed it is identical to the human annexin I (Anx-I) protein. The identity of the isolated P-40 cDNA as Anx-I was confirmed by the specific binding of IPM96 mAb to a 40kDa protein following the in vitro expression of P-40 full-length cDNA. Northern blot analysis of total RNA from drug-sensitive and -resistant cells revealed an increase in P-40 (or Anx-I) mRNA in drug-resistant cells relative to drug-sensitive cells. Transfection of Anx-I cDNA into drug-sensitive MCF-7 cells was carried out without further drug selection and showed 2- to 5-fold increase in resistance of transfected cells to adriamycin, melphalan, and etoposide. Conversely, transfection of reverse Anx-I cDNA into SKOV-3 cells decreased the expression of Anx-I without affecting the expression of other members of the annexin family and showed a 3- to 8-fold increase in sensitivity to these drugs. Of interest was the correlation between the presence of Anx-I and MDR in MDA-MB-231 cells when compared to MCF-7 cells. MDA-MB-231 cells show 3- to 20-fold increase in resistance to adriamycin, melphalan, and etoposide in the absence of detectable levels of P-glycoprotein (P-gp1), the multidrug resistance protein (MRP1) or the breast cancer resistance protein (BCRP). Taken together, these results provide the first direct evidence for the role of Anx-I in MDR of tumor cells.