Clinical drug response prediction from preclinical cancer cell lines by logistic matrix factorization approach

Predicting tumor drug response using cancer cell line drug response values for a large number of anti-cancer drugs is a significant challenge in personalized medicine. Predicting patient response to drugs from data obtained from preclinical models is made easier by the availability of different knowledge on cell lines and drugs. This paper proposes the TCLMF method, a predictive model for predicting drug response in tumor samples that was trained on preclinical samples and is based on the logistic matrix factorization approach. The TCLMF model is designed based on gene expression profiles, tissue type information, the chemical structure of drugs and drug sensitivity (IC 50) data from cancer cell lines. We use preclinical data from the Genomics of Drug Sensitivity in Cancer dataset (GDSC) to train the proposed drug response model, which we then use to predict drug sensitivity of samples from the Cancer Genome Atlas (TCGA) dataset. The TCLMF approach focuses on identifying successful features of cell lines and drugs in order to calculate the probability of the tumor samples being sensitive to drugs. The closest cell line neighbours for each tumor sample are calculated using a description of similarity between tumor samples and cell lines in this study. The drug response for a new tumor is then calculated by averaging the low-rank features obtained from its neighboring cell lines. We compare the results of the TCLMF model with the results of the previously proposed methods using two databases and two approaches to test the model's performance. In the first approach, 12 drugs with enough known clinical drug response, considered in previous methods, are studied. For 7 drugs out of 12, the TCLMF can significantly distinguish between patients that are resistance to these drugs and the patients that are sensitive to them. These approaches are converted to classification models using a threshold in the second approach, and the results are compared. The results demonstrate that the TCLMF method provides accurate predictions across the results of the other algorithms. Finally, we accurately classify tumor tissue type using the latent vectors obtained from TCLMF's logistic matrix factorization process. These findings demonstrate that the TCLMF approach produces effective latent vectors for tumor samples. The source code of the TCLMF method is available in https://github.com/emdadi/TCLMF.

Bleomycin Concentration in Patients' Plasma and Tumors after Electrochemotherapy. A Study from InspECT Group

The plasma concentration profile of bleomycin in the distribution phase of patients younger than 65 years is needed to determine the suitable time interval for efficient application of electric pulses during electrochemotherapy. Additionally, bleomycin concentrations in the treated tumors for effective tumor response are not known. In this study, the pharmacokinetic profile of bleomycin in the distribution phase in 12 patients younger than 65 years was determined. In 17 patients, the intratumoral bleomycin concentration was determined before the application of electric pulses. In younger patients, the pharmacokinetics of intravenously injected bleomycin demonstrated a faster plasma clearance rate than that in patients older than 65 years. This outcome might indicate that the lowering of the standard bleomycin dose of 15,000 IU/m² with intravenous bleomycin injection for electrochemotherapy is not recommended in younger patients. Based on the plasma concentration data gathered, a time interval for electrochemotherapy of 5-15 min after bleomycin injection was determined. The median bleomycin concentration in tumors 8 min after bleomycin injection, at the time of electroporation, was 170 ng/g. Based on collected data, the reduction of the bleomycin dose is not recommended in younger patients; however, a shortened time interval for application of electric pulses in electrochemotherapy to 5-15 min after intravenous bleomycin injection should be considered.

Comparison of the Allelic Alterations between InDel and STR Markers in Tumoral Tissues Used for Forensic Purposes

Background and objectives: Over the last two decades, human DNA identification and kinship tests have been conducted mainly through the analysis of short tandem repeats (STRs). However, other types of markers, such as insertion/deletion polymorphisms (InDels), may be required when DNA is highly degraded. In forensic genetics, tumor samples may sometimes be used in some cases of human DNA identification and in paternity tests. Nevertheless, tumor genomic instability related to forensic DNA markers should be considered in forensic analyses since it can compromise genotype attribution. Therefore, it is useful to know what impact tumor transformation may have on the forensic interpretation of the results obtained from the analysis of these polymorphisms. Materials and Methods: The aim of this study was to investigate the genomic instability of InDels and STRs through the analysis of 55 markers in healthy tissue and tumor samples (hepatic, gastric, breast, and colorectal cancer) in 66 patients. The evaluation of genomic instability was performed comparing InDel and STR genotypes of tumor samples with those of their healthy counterparts. Results: With regard to STRs, colorectal cancer was found to be the tumor type affected by the highest number of mutations, whereas in the case of InDels the amount of genetic mutations turned out to be independent of the tumor type. However, the phenomena of genomic instability, such as loss of heterozygosity (LOH) and microsatellite instability (MSI), seem to affect InDels more than STRs hampering genotype attribution. Conclusion: We suggest that the use of STRs rather than InDels could be more suitable in forensic genotyping analyses given that InDels seem to be more affected than STRs by mutation events capable of compromising genotype attribution.

Use of archival versus newly collected tumor samples for assessing PD-L1 expression and overall survival: an updated analysis of KEYNOTE-010 trial

BACKGROUND

In KEYNOTE-010, pembrolizumab versus docetaxel improved overall survival (OS) in patients with programmed death-1 protein (PD)-L1-positive advanced non-small-cell lung cancer (NSCLC). A prespecified exploratory analysis compared outcomes in patients based on PD-L1 expression in archival versus newly collected tumor samples using recently updated survival data.

PATIENTS AND METHODS

PD-L1 was assessed centrally by immunohistochemistry (22C3 antibody) in archival or newly collected tumor samples. Patients received pembrolizumab 2 or 10 mg/kg Q3W or docetaxel 75 mg/m2 Q3W for 24 months or until progression/intolerable toxicity/other reason. Response was assessed by RECIST v1.1 every 9 weeks, survival every 2 months. Primary end points were OS and progression-free survival (PFS) in tumor proportion score (TPS) ≥50% and ≥1%; pembrolizumab doses were pooled in this analysis.

RESULTS

At date cut-off of 24 March 2017, median follow-up was 31 months (range 23-41) representing 18 additional months of follow-up from the primary analysis. Pembrolizumab versus docetaxel continued to improve OS in patients with previously treated, PD-L1-expressing advanced NSCLC; hazard ratio (HR) was 0.66 [95% confidence interval (CI): 0.57, 0.77]. Of 1033 patients analyzed, 455(44%) were enrolled based on archival samples and 578 (56%) on newly collected tumor samples. Approximately 40% of archival samples and 45% of newly collected tumor samples were PD-L1 TPS ≥50%. For TPS ≥50%, the OS HRs were 0.64 (95% CI: 0.45, 0.91) and 0.40 (95% CI: 0.28, 0.56) for archival and newly collected samples, respectively. In patients with TPS ≥1%, OS HRs were 0.74 (95% CI: 0.59, 0.93) and 0.59 (95% CI: 0.48, 0.73) for archival and newly collected samples, respectively. In TPS ≥50%, PFS HRs were similar across archival [0.63 (95% CI: 0.45, 0.89)] and newly collected samples [0.53 (95% CI: 0.38, 0.72)]. In patients with TPS ≥1%, PFS HRs were similar across archival [0.82 (95% CI: 0.66, 1.02)] and newly collected samples [0.83 (95% CI: 0.68, 1.02)].

CONCLUSION

Pembrolizumab continued to improve OS over docetaxel in intention to treat population and in subsets of patients with newly collected and archival samples.

TRIAL REGISTRATION

ClinicalTrials.gov: NCT01905657.