携带同源重组修复相关基因突变的消化道肿瘤的临床特点

目的:探讨消化道肿瘤中同源重组修复相关基因(homologous recombination repair related gene，HRR)突变的发生情况及临床意义。方法:共92例消化道肿瘤患者，79例患者进行了血液标本HRR检测，53例患者进行了组织标本HRR检测，40例患者同时行血液和组织的HRR基因检测，收集患者基因检测结果及临床相关资料。结果:在79例患者血液标本检测中发现10例（12.6%）有临床意义HRR突变，在53例患者组织标本检测中发现9例（17.0%）有临床意义HRR突变。40例同时行血液和组织的HRR基因检测患者中常见的有临床意义HRR突变为CDK12突变4例（10.0%）、ATM突变3例（7.5%）、BRCA1突变2例（5.0%）。13例有临床意义HRR突变患者中常见共存突变为TP53突变10例（76.9%）、APC突变5例（38.5%）、PIK3CA突变4例（30.8%）。40例患者中13例患者血液和/或组织中有临床意义HRR突变，27例患者血液和组织中均无任何临床意义HRR突变且两组相比，有临床意义HRR突变组肿瘤突变负荷（tumor mutational burden，TMB）为6.17(2.24～11.52)，而未携带HRR突变组TMB为0.4(0～3.75)，差异有统计学意义（P＜0.05）。40例患者组织检测中7例HRR有临床意义的突变，33例无HRR突变，血液检测中10例HRR有临床意义的突变，30例无HRR突变，一致性检验的Kappa值为0.333（P=0.031）。结论:携带有临床意义HRR突变的消化道肿瘤患者TMB更高，血液和组织检测HRR突变有较好的一致性。     

Skin3D: Detection and longitudinal tracking of pigmented skin lesions in 3D total-body textured meshes

We present an automated approach to detect and longitudinally track skin lesions on 3D total-body skin surface scans. The acquired 3D mesh of the subject is unwrapped to a 2D texture image, where a trained objected detection model, Faster R-CNN, localizes the lesions within the 2D domain. These detected skin lesions are mapped back to the 3D surface of the subject and, for subjects imaged multiple times, we construct a graph-based matching procedure to longitudinally track lesions that considers the anatomical correspondences among pairs of meshes and the geodesic proximity of corresponding lesions and the inter-lesion geodesic distances.We evaluated the proposed approach using 3DBodyTex, a publicly available dataset composed of 3D scans imaging the coloured skin (textured meshes) of 200 human subjects. We manually annotated locations that appeared to the human eye to contain a pigmented skin lesion as well as tracked a subset of lesions occurring on the same subject imaged in different poses. Our results, when compared to three human annotators, suggest that the trained Faster R-CNN detects lesions at a similar performance level as the human annotators. Our lesion tracking algorithm achieves an average matching accuracy of 88% on a set of detected corresponding pairs of prominent lesions of subjects imaged in different poses, and an average longitudinal accuracy of 71% when encompassing additional errors due to lesion detection. As there currently is no other large-scale publicly available dataset of 3D total-body skin lesions, we publicly release over 25,000 3DBodyTex manual annotations, which we hope will further research on total-body skin lesion analysis.     

Pharmacokinetic study of the oral fluorouracil antitumor agent S‐1 in patients with impaired renal function

Although dose reduction of S‐1 is recommended for patients with impaired renal function, dose modification for such patients has not been prospectively evaluated. The aim of the present study was to investigate the pharmacokinetic parameters of 5‐fluorouracil, 5‐chloro‐2,4 dihydroxypyridine and oteracil potassium, and to review the recommended dose modification of S‐1 in patients with renal impairment. We classified patients receiving S‐1 into 4 groups according to their renal function, as measured using the Japanese estimated glomerular filtration rate (eGFR) equation. The daily S‐1 dose was adjusted based on the patient's eGFR and body surface area. Blood samples were collected for pharmacokinetic analysis. A total of 33 patients were enrolled and classified into 4 groups as follows: 10 patients in cohort 1 (eGFR ≥ 80 mL/min/1.73 m²), 10 patients in cohort 2 (eGFR = 50‐79 mL/min/1.73 m²), 10 patients in cohort 3 (eGFR = 30‐49 mL/min/1.73 m²), and 3 patients in cohort 4 (eGFR < 30 mL/min/1.73 m²). Those in cohorts 3 and 4 treated with an adjusted dose of S‐1 showed a similar area under the curve for 5‐fluorouracil (941.9 ± 275.6 and 1043.5 ± 224.8 ng/mL, respectively) compared with cohort 2 (1034.9 ± 414.3 ng/mL). Notably, while there was a statistically significant difference between cohort 1 (689.6 ± 208.8 ng/mL) and 2 (P = 0.0474) treated with an equal dose of S‐1, there was no significant difference observed in the toxicity profiles of the cohorts. In conclusion, dose adjustment of S‐1 in patients with impaired renal function using eGFR is appropriate and safe.     

Efficacy and safety of pembrolizumab for the treatment of advanced biliary cancer: Results from the KEYNOTE-158 and KEYNOTE-028 studies

We present data from patients with advanced biliary tract cancer (BTC) receiving pembrolizumab in the KEYNOTE-158 (NCT02628067; phase 2) and KEYNOTE-028 (NCT02054806; phase 1b) studies. Eligible patients aged ≥18 years from both studies had histologically/cytologically confirmed incurable BTC that progressed after standard treatment regimen(s), measurable disease per Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1, Eastern Cooperative Oncology Group performance status 0/1, and no prior immunotherapy. Programmed death ligand 1 (PD-L1)-positive tumors were required for eligibility in KEYNOTE-028 only. Patients received pembrolizumab 200 mg every three weeks (KEYNOTE-158) or 10 mg/kg every two weeks (KEYNOTE-028) for ≤2 years. Primary efficacy endpoint was objective response rate (ORR) by RECIST v1.1. Response assessed by independent central review is reported. KEYNOTE-158 enrolled 104 patients and KEYNOTE-028 enrolled 24 patients. Median (range) follow-up was 7.5 months (0.6-34.3) in KEYNOTE-158 and 5.7 months (0.6-55.4) in KEYNOTE-028. In KEYNOTE-158, ORR was 5.8% (6/104; 95% CI, 2.1%-12.1%); median duration of response (DOR) was not reached (NR) (range, 6.2-26.6+ months). Median (95% CI) OS and PFS were 7.4 (5.5-9.6) and 2.0 (1.9-2.1) months. Among PD-L1-expressers (n = 61) and PD-L1-nonexpressers (n = 34), respectively, ORR was 6.6% (4/61) and 2.9% (1/34). In KEYNOTE-028, ORR was 13.0% (3/23; 95% CI, 2.8%-33.6%); median DOR was NR (range, 21.5-53.2+ months). Median (95% CI) OS and PFS were 5.7 (3.1-9.8) and 1.8 (1.4-3.1) months. Grade 3 to 5 treatment-related adverse events occurred in 13.5% of patients in KEYNOTE-158 (no grade 4; grade 5 renal failure, n = 1) and 16.7% in KEYNOTE-028 (no grade 4/5). In summary, pembrolizumab provides durable antitumor activity in 6% to 13% of patients with advanced BTC, regardless of PD-L1 expression, and has manageable toxicity.     

Can natural language processing provide accurate,automated reporting of wound infection requiring reoperation after lumbar discectomy?

BACKGROUNDSurgical site infections are a major driver of morbidity and increased costs in the postoperative period after spine surgery. Current tools for surveillance of these adverse events rely on prospective clinical tracking, manual retrospective chart review, or administrative procedural and diagnosis codes.PURPOSEThe purpose of this study was to develop natural language processing (NLP) algorithms for automated reporting of postoperative wound infection requiring reoperation after lumbar discectomy.PATIENT SAMPLEAdult patients undergoing discectomy at two academic and three community medical centers between January 1, 2000 and July 31, 2019 for lumbar disc herniation.OUTCOME MEASURESReoperation for wound infection within 90 days after surgeryMETHODSFree-text notes of patients who underwent surgery from January 1, 2000 to December 31, 2015 were used for algorithm training. Free-text notes of patients who underwent surgery after January 1, 2016 were used for algorithm testing. Manual chart review was used to label which patients had reoperation for wound infection. An extreme gradient-boosting NLP algorithm was developed to detect reoperation for postoperative wound infection.RESULTSOverall, 5,860 patients were included in this study and 62 (1.1%) had a reoperation for wound infection. In patients who underwent surgery after January 1, 2016 (n=1,377), the NLP algorithm detected 15 of the 16 patients (sensitivity=0.94) who had reoperation for infection. In comparison, current procedural terminology and international classification of disease codes detected 12 of these 16 patients (sensitivity=0.75). At a threshold of 0.05, the NLP algorithm had positive predictive value of 0.83 and F1-score of 0.88.CONCLUSIONTemporal validation of the algorithm developed in this study demonstrates a proof-of-concept application of NLP for automated reporting of adverse events after spine surgery. Adapting this methodology for other procedures and outcomes in spine and orthopedics has the potential to dramatically improve and automatize quality and safety reporting.