Prediction model for major bleeding in anticoagulated patients with cancer-associated venous thromboembolism using machine learning and natural language processing.

PURPOSE: We developed a predictive model to assess the risk of major bleeding (MB) within 6 months of primary venous thromboembolism (VTE) in cancer patients receiving anticoagulant treatment. We also sought to describe the prevalence and incidence of VTE in cancer patients, and to describe clinical characteristics at baseline and bleeding events during follow-up in patients receiving anticoagulants. METHODS: This observational, retrospective, and multicenter study used natural language processing and machine learning (ML), to analyze unstructured clinical data from electronic health records from nine Spanish hospitals between 2014 and 2018. All adult cancer patients with VTE receiving anticoagulants were included. Both clinically- and ML-driven feature selection was performed to identify MB predictors. Logistic regression (LR), decision tree (DT), and random forest (RF) algorithms were used to train predictive models, which were validated in a hold-out dataset and compared to the previously developed CAT-BLEED score. RESULTS: Of the 2,893,108 cancer patients screened, in-hospital VTE prevalence was 5.8% and the annual incidence ranged from 2.7 to 3.9%. We identified 21,227 patients with active cancer and VTE receiving anticoagulants (53.9% men, median age of 70 years). MB events after VTE diagnosis occurred in 10.9% of patients within the first six months. MB predictors included: hemoglobin, metastasis, age, platelets, leukocytes, and serum creatinine. The LR, DT, and RF models had AUC-ROC (95% confidence interval) values of 0.60 (0.55, 0.65), 0.60 (0.55, 0.65), and 0.61 (0.56, 0.66), respectively. These models outperformed the CAT-BLEED score with values of 0.53 (0.48, 0.59). CONCLUSIONS: Our study shows encouraging results in identifying anticoagulated patients with cancer-associated VTE who are at high risk of MB.

Melatonin Activates Endoplasmic Reticulum Stress and Apoptosis in Rats with Diethylnitrosamine-Induced Hepatocarcinogenesis.

Hepatocellular carcinoma (HCC) is one of the most lethal human cancers worldwide because of its high incidence, its metastatic potential and the low efficacy of conventional treatment. Inactivation of apoptosis is implicated in tumour progression and chemotherapy resistance, and has been linked to the presence of endoplasmic reticulum stress. Melatonin, the main product of the pineal gland, exerts anti-proliferative, pro-apoptotic and anti-angiogenic effects in HCC cells, but these effects still need to be confirmed in animal models. Male Wistar rats in treatment groups received diethylnitrosamine (DEN) 50 mg/kg intraperitoneally twice/once a week for 18 weeks. Melatonin was given in drinking water at 1 mg/kg/d, beginning 5 or 12 weeks after the start of DEN administration. Melatonin improved survival rates and successfully attenuated liver injury, as shown by histopathology, decreased levels of serum transaminases and reduced expression of placental glutathione S-transferase. Furthermore, melatonin treatment resulted in a significant increase of caspase 3, 8 and 9 activities, polyadenosine diphosphate (ADP) ribose polymerase (PARP) cleavage, and Bcl-associated X protein (Bax)/Bcl-2 ratio. Cytochrome c, p53 and Fas-L protein concentration were also significantly enhanced by melatonin. Melatonin induced an increased expression of activating transcription factor 6 (ATF6), C/EBP-homologous protein (CHOP) and immunoglobulin heavy chain-binding protein (BiP), while cyclooxygenase (COX)-2 expression decreased. Data obtained suggest that induction of apoptosis and ER stress contribute to the beneficial effects of melatonin in rats with DEN-induced HCC.

A dose-dense schedule of docetaxel followed by doxorubicin and cyclophosphamide as neoadjuvant treatment for breast cancer: results from a phase II study.

INTRODUCTION: The objective was to evaluate a dose-dense schedule of docetaxel followed by doxorubicin and cyclophosphamide (AC) as neoadjuvant treatment for patients with locally advanced breast cancer. PATIENTS AND METHODS: Ninety-nine patients were included and received 100 mg/m(2) of docetaxel every two weeks for four cycles followed by 60 mg/m(2) of doxorubicin and 600 mg/m(2) of cyclophosphamide every two weeks for four cycles. Primary prophylaxis with granulocyte colony-stimulating factor (G-CSF) was administered systematically to all patients. RESULTS: Efficacy and toxicity analyses were carried out on an intention-to-treat basis. After treatment, complete pathological response in the breast and lymph nodes was confirmed in 15 patients (15%, 95% confidence interval [CI]: 8.4-22.9). Clinical response rate was 74% (95% CI: 65-82), of which 19% were complete responses. Breast-conserving surgery could be performed in 41% of patients. The dose-dense schedule was generally well tolerated. The most important grade 3/4 toxicities per patient were cutaneous toxicity (12.1%) and hepatic dysfunction (9.1%) during docetaxel administration, and neutropenia (28.1%) and leucopenia (8.3%) with AC. CONCLUSION: A dose-dense schedule of docetaxel followed by AC as neoadjuvant treatment is an effective and safe treatment for locally advanced breast cancer. Primary prophylaxis with G-CSF, and possibly the change in the sequence of drug administration, appears to play a major role in avoiding the excessive toxicity of dose-dense schedules.

Phase II study of dose-dense doxorubicin and docetaxel as neoadjunvant chemotherapy with G-CSF support in patients with large or locally advanced breast cancer.

INTRODUCTION: To evaluate the efficacy and safety profile of the concomitant dose-dense administration of doxorubicin and docetaxel as primary chemotherapy for patients with large or locally advanced breast cancer. MATERIALS AND METHODS: Forty-seven patients were included and received 50 mg/m(2) of doxorubicin and 75 mg/m(2) of docetaxel every two weeks for four cycles. Primary prophylaxis with granulocyte colony stimulating factor was administered. RESULTS: Patients included had mainly stage III disease (66%). Efficacy and toxicity analyses were carried out on an intention-to-treat basis. After study treatment, the rate of clinical responses was 85% (95% CI: 75-95) with 6% judged as clinical complete responses. Surgery was performed on 94% patients for whom the breast was conserved in 27%. Only one patient obtained a pathological complete response (with no evidence of invasive or non-invasive tumour in the breast and the lymph nodes). In three additional patients, malignant cells were detected only in one lymph node. The single severe haematological toxicity was neutropenia, occurring in one patient (2%) and two cycles (1%), being grade 3 in one and grade 4 in the other. Severe non-haematological toxicities were grade 3, and the most common was asthenia (8% of patients), followed by cutaneous toxicity, arthromyalgia and stomatitis, which occurred in fewer than 4% of patients in each case. CONCLUSIONS: The concomitant dose-dense administration of doxorubicin and docetaxel as neoadjuvant chemotherapy with granulocyte colony stimulating factor support is a feasible and effective schedule with a safe toxicity profile for women with large or locally advanced breast cancer.

Phase II trial of fortnightly irinotecan (CPT-11) in the treatment of colorectal cancer patients resistant to previous fluoropyrimidine-based chemotherapy.

INTRODUCTION: This phase II study investigated the anti-tumour activity and toxicity of CPT-11 (250 mg/m2 i.v. infusion over 60 minutes) administered every 2 weeks as second-line chemotherapy in patients with advanced colorectal cancer (CRC). MATERIAL AND METHODS: Patients (n = 63) with histology diagnosis of advanced CRC and proven resistance to previous fluoropyrimidine therapy were enrolled. RESULTS: A total of 510 CPT-11 cycles were administered, with a mean of 8 cycles per patient (range: 1-32). The median relative dose intensity was 93%. Partial response (PR) was obtained in 11 patients (17.5%; 95%CI: 8.1%-26.7%) and 29 patients (46.0%) showed stable disease (clinical benefit of 63.5%). The median duration of response was 6.8 months (95%CI: 6.1-7.5 months), median survival was 8.8 months (95%CI: 6.3-11.5 months) and median time to disease progression was 4.5 months (95%CI: 3.9-5.0 months). Overall, this schedule of CPT-11 chemotherapy was well tolerated by the patient. Neutropenia was the most frequent grade 3/4 haematological toxicity (20.6% of patients and 4.1% of cycles). Neutropenia with concurrent fever or infection occurred in 7 patients (11.1%). Late onset diarrhoea was the most frequent grade 3/4 non-haematological toxicity (19.0% of patients and 2.3% of cycles). Other, lower-incidence, toxicities were anaemia, fever, infection, mucositis, nausea and vomiting. There were no toxic deaths. CONCLUSIONS: We found that CPT-11, administered as 250 mg/m2 i.v. infusion over 60 minutes every 2 weeks, was active and well tolerated schedule in the second-line chemotherapy of advanced CRC patients. This bi-weekly scheme could be used as an alternative to the weekly or the every-three-week schedule as well as in combined therapies with other chemotherapeutic agents for the treatment of advanced, metastatic, CRC.