Early Diagnosis of Pancreatic Cancer via Machine Learning Analysis of a National Electronic Medical Record Database.

PURPOSE: Pancreatic cancer (PaC) is often diagnosed at advanced stages, resulting in one of the lowest survival rates among patients with cancer. The purpose of this study was to investigate whether machine learning (ML) models can predict with high sensitivity and specificity an increased risk for PaC ahead of clinical diagnosis. METHODS: Optum deidentified electronic health record (EHR) data set was used to extract 1-year data for each patient and to sample for PaC diagnosis, the number of interactions with the health care system, and unique demographic and clinical features. Data for patients with PaC diagnosis were collected between 1 and 2 years before the diagnosis. Standard binary classification ML models were used on training and testing data sets. Data analyses were performed using the scikit-learn package version 1.0.1. RESULTS: The data set consisted of 18,987 patient EHRs collected between December 31, 2007, and December 31, 2017. EHRs with 10 unique features and at least three health care interactions were used for model training (N = 15,189; n = 8,438 [56%] with PaC) and testing (N = 3,798; n = 2,127 [56%] with PaC). The ensemble model achieved an AUC of 0.89, a sensitivity of 85.61%, and a specificity of 76.18% on the testing data set and produced superior results compared with other binary classifiers. Increasing unique health care interactions to nine failed to improve the AUC score. When the testing data set was enlarged to 5,696 patients, the ensemble model achieved an AUC of 0.92 and a specificity of 93.21%, but the sensitivity was compromised. CONCLUSION: The ensemble model exceeded the state-of-the-art level of performance for prediction of PaC ahead of clinical diagnosis with a minimal clinically guided input, providing a potential strategy for selection of high-risk patients for further screening.

Long-term treatment-free remission in patients with chronic myeloid leukemia after second-line nilotinib: ENESTop 5-year update.

The ENESTop study evaluated treatment-free remission (TFR) in patients with chronic myeloid leukemia (CML) in chronic phase who had received ≥3 years of tyrosine kinase inhibitor therapy and achieved sustained deep molecular response only after switching from imatinib to nilotinib. After 1-year nilotinib consolidation, 126 patients attempted TFR. At 48 weeks (primary analysis), 57.9% (73/126) were in TFR. In the present analysis at 5 years, 42.9% (54/126) were in TFR. Since the 48-week analysis, among patients who left the TFR phase, 58% (11/19) did not have a loss of molecular response and discontinued for other reasons. Of the 59 patients who reinitiated nilotinib upon loss of major molecular response (MMR) or confirmed loss of MR4, 98.3% regained MMR, 94.9% regained MR4, and 93.2% regained MR4.5. Overall adverse event rates decreased over the 5 years of TFR. In patients reinitiating nilotinib, there was a cumulative increase in cardiovascular events with longer nilotinib exposure. No disease progression or CML-related deaths were reported. Overall, these results confirm the durability and safety of TFR for patients receiving second-line nilotinib. Cardiovascular risk should be carefully managed, particularly when reinitiating treatment after TFR.

Efficacy and safety of ofatumumab and bendamustine followed by ofatumumab maintenance in patients with relapsed indolent non-Hodgkin lymphoma after prior rituximab.

In indolent non-Hodgkin's lymphoma (iNHL), patients treated with rituximab, alone or in combination with various chemotherapeutic agents eventually relapse. This study evaluated the combination of ofatumumab and bendamustine, followed by maintenance ofatumumab in patients with relapsed iNHL with prior sensitivity to rituximab. Among the 49 patients enrolled, 24.5% achieved a complete response (CR) and 42.9% achieved a partial response (PR), with an overall response rate of 67.3% at the end of the induction therapy. Additionally, six patients with PR during induction phase achieved CR during the maintenance phase. Treatment-related adverse event was observed in 95.9% patients. The most common hematologic and biochemical abnormalities were decrease in lymphocytes (85.7%) and increase in glucose (91.8%), respectively. Overall, 42.9% progressed and 14.3% died during the study. Thus, ofatumumab in combination with bendamustine, followed by ofatumumab maintenance, was effective in the treatment of patients with iNHL with a manageable safety profile (NCT01294579).

Treatment-Free Remission After Second-Line Nilotinib Treatment in Patients With Chronic Myeloid Leukemia in Chronic Phase: Results From a Single-Group, Phase 2, Open-Label Study.

Background: Treatment-free remission (TFR)-that is, stopping tyrosine kinase inhibitor (TKI) therapy without loss of response-is an emerging treatment goal in chronic myeloid leukemia (CML). Objective: To evaluate TFR after discontinuation of second-line nilotinib therapy. Design: Single-group, phase 2, open-label study. (ClinicalTrials.gov: NCT01698905). Setting: 63 centers in 18 countries. Patients: Adults with CML in chronic phase who received TKI therapy for at least 3 years (>4 weeks with imatinib, then ≥2 years with nilotinib) and achieved MR4.5 (BCR-ABL1 ≤0.0032% on the International Scale [BCR-ABL1IS]) while receiving nilotinib entered a 1-year consolidation phase. Those with sustained MR4.5 during consolidation were eligible to enter TFR. Interventions: Patients received nilotinib during consolidation; those who entered TFR stopped treatment. Patients with loss of major molecular response (MMR) (BCR-ABL1IS ≤0.1%) or confirmed loss of MR4 (BCR-ABL1IS ≤0.01%) during TFR reinitiated nilotinib treatment. Measurements: Proportion of patients without loss of MMR, confirmed loss of MR4, or treatment reinitiation within 48 weeks of stopping treatment (primary end point). Results: 163 patients who had switched from imatinib to nilotinib (for reasons including resistance, intolerance, and physician preference) enrolled in the study and entered the consolidation phase. Of these patients, 126 met the criteria for entering the TFR phase, and 73 (58% [95% CI, 49% to 67%]) and 67 (53% [CI, 44% to 62%]) maintained TFR at 48 weeks (primary end point) and 96 weeks, respectively. Of the 56 patients who reinitiated nilotinib therapy, 55 regained MMR or better and 52 regained MR4.5. None had CML progression to accelerated phase or blast crisis. Musculoskeletal pain was more frequent during the first 48 weeks after nilotinib discontinuation. Limitation: The study included a heterogeneous patient population and was not designed to compare outcomes between patients continuing and those stopping treatment. Conclusion: TFR seems achievable in patients with sustained MR4.5 after switching to nilotinib. Primary Funding Source: Novartis Pharmaceuticals Corporation.

Switching to nilotinib versus imatinib dose escalation in patients with chronic myeloid leukaemia in chronic phase with suboptimal response to imatinib (LASOR): a randomised, open-label trial.

BACKGROUND: Optimal management of patients with chronic myeloid leukaemia in chronic phase with suboptimal cytogenetic response remains undetermined. This study aimed to investigate the safety and efficacy of switching to nilotinib vs imatinib dose escalation for patients with suboptimal cytogenetic response on imatinib. METHODS: We did a phase 3, open-label, randomised trial in patients with chronic myeloid leukaemia in chronic phase with suboptimal cytogenetic response to imatinib according to the 2009 European LeukemiaNet criteria, in Latin America, Europe, and Asia (59 hospitals and care centres in 12 countries). Eligible patients were aged 18 years or older with Philadelphia chromosome-positive chronic myeloid leukaemia in chronic phase and Eastern Cooperative Oncology Group performance status of 0-2. Before enrolment, all patients had received 3-18 months of imatinib 400 mg once daily and had a suboptimal cytogenetic response according to 2009 ELN recommendations, established through bone marrow cytogenetics. By use of an interactive response technology using fixed blocks, we randomly assigned patients (1:1) to switch to nilotinib 400 mg twice per day or an escalation of imatinib dose to 600 mg once per day (block size of 4). Investigators and participants were not blinded to study treatment. Crossover was allowed for loss of response or intolerance at any time, or for patients with no complete cytogenetic response at 6 months. The primary endpoint was complete cytogenetic response at 6 months in the intention-to-treat population. Efficacy endpoints were based on the intention-to-treat population, with all patients assessed according to the treatment group to which they were randomised (regardless of crossover); the effect of crossover was assessed in post-hoc analyses, in which responses achieved after crossover were excluded. We present the final results at 24 months' follow-up. This study is registered with ClinicalTrials.gov (NCT00802841). FINDINGS: Between July 7, 2009, and Aug 29, 2012, we enrolled 191 patients. 96 patients were randomly assigned to nilotinib and 95 patients were randomly assigned to imatinib. Complete cytogenetic response at 6 months was achieved by 48 of 96 patients in the nilotinib group (50%, 95·18% CI 40-61) and 40 of 95 in the imatinib group (42%, 32-53%; difference 7·9% in favour of nilotinib; 95% CI -6·2 to 22·0, p=0·31). Excluding responses achieved after crossover, 48 (50%) of 96 patients in the nilotinib group and 34 (36%) of 95 patients in the imatinib group achieved complete cytogenic response at 6 months (nominal p=0·058). Grade 3-4 non-haematological adverse events occurring in more than one patient were headache (nilotinib group, n=2 [2%, including 1 after crossover to imatinib]; imatinib group, n=1 [1%]), blast cell crisis (nilotinib group, n=1 [1%]; imatinib group, n=1 [1%]), and QT prolongation (nilotinib group, n=1 [1%]; imatinib group, n=1 [1%, after crossover to nilotinib]). Serious adverse events on assigned treatment were reported in 11 (11%) of 96 patients in the nilotinib group and nine (10%) of 93 patients in the imatinib group. Seven (7%) of 96 patients died in the nilotinib group and five (5%) of 93 patients died in the imatinib group; no deaths were treatment-related. INTERPRETATION: While longer-term analyses are needed to establish whether the clinical benefits observed with switching to nilotinib are associated with improved long-term survival outcomes, our results suggest that patients with suboptimal cytogenetic response are more likely to achieve improved cytogenetic and molecular responses with switching to nilotinib than with imatinib dose escalation, although the difference was not statistically significant when responses achieved after crossover were included. FUNDING: Novartis Pharmaceuticals.