Promising Candidate Prognostic Biomarkers in [18F]FDG PET Images: Evaluation in Independent Cohorts of Non-Small Cell Lung Cancer Patients.

The normalized distances from the hot spot of radiotracer uptake (SUVmax) to the tumor centroid (NHOC) and to the tumor perimeter (NHOP) have recently been suggested as novel PET features reflecting tumor aggressiveness. These biomarkers characterizing the shift of SUVmax toward the lesion edge during tumor progression have been shown to be prognostic factors in breast and non-small cell lung cancer (NSCLC) patients. We assessed the impact of imaging parameters on NHOC and NHOP, their complementarity to conventional PET features, and their prognostic value for advanced-NSCLC patients. Methods: This retrospective study investigated baseline [18F]FDG PET scans: cohort 1 included 99 NSCLC patients with no treatment-related inclusion criteria (robustness study); cohort 2 included 244 NSCLC patients (survival analysis) treated with targeted therapy (93), immunotherapy (63), or immunochemotherapy (88). Although 98% of patients had metastases, radiomic features including SUVs were extracted from the primary tumor only. NHOCs and NHOPs were computed using 2 approaches: the normalized distance from the localization of SUVmax or SUVpeak to the tumor centroid or perimeter. Bland-Altman analyses were performed to investigate the impact of both spatial resolution (comparing PET images with and without gaussian postfiltering) and image sampling (comparing 2 voxel sizes) on feature values. The correlation of NHOCs and NHOPs with other features was studied using Spearman correlation coefficients (r). The ability of NHOCs and NHOPs to predict overall survival (OS) was estimated using the Kaplan-Meier method. Results: In cohort 1, NHOC and NHOP features were more robust to image filtering and to resampling than were SUVs. The correlations were weak between NHOCs and NHOPs (r ≤ 0.45) and between NHOCs or NHOPs and any other radiomic features (r ≤ 0.60). In cohort 2, the patients with short OS demonstrated higher NHOCs and lower NHOPs than those with long OS. NHOCs significantly distinguished 2 survival profiles in patients treated with immunotherapy (log-rank test, P < 0.01), whereas NHOPs stratified patients regarding OS in the targeted therapy (P = 0.02) and immunotherapy (P < 0.01) subcohorts. Conclusion: Our findings suggest that even in advanced NSCLC patients, NHOC and NHOP features pertaining to the primary tumor have prognostic potential. Moreover, these features appeared to be robust with respect to imaging protocol parameters and complementary to other radiomic features and are now available in LIFEx software to be independently tested by others.

Consore: A Powerful Federated Data Mining Tool Driving a French Research Network to Accelerate Cancer Research.

BACKGROUND: Real-world data (RWD) related to the health status and care of cancer patients reflect the ongoing medical practice, and their analysis yields essential real-world evidence. Advanced information technologies are vital for their collection, qualification, and reuse in research projects. METHODS: UNICANCER, the French federation of comprehensive cancer centres, has innovated a unique research network: Consore. This potent federated tool enables the analysis of data from millions of cancer patients across eleven French hospitals. RESULTS: Currently operational within eleven French cancer centres, Consore employs natural language processing to structure the therapeutic management data of approximately 1.3 million cancer patients. These data originate from their electronic medical records, encompassing about 65 million medical records. Thanks to the structured data, which are harmonized within a common data model, and its federated search tool, Consore can create patient cohorts based on patient or tumor characteristics, and treatment modalities. This ability to derive larger cohorts is particularly attractive when studying rare cancers. CONCLUSIONS: Consore serves as a tremendous data mining instrument that propels French cancer centres into the big data era. With its federated technical architecture and unique shared data model, Consore facilitates compliance with regulations and acceleration of cancer research projects.

How to Improve Cancer Patients ENrollment in Clinical Trials From rEal-Life Databases Using the Observational Medical Outcomes Partnership Oncology Extension: Results of the PENELOPE Initiative in Urologic Cancers.

PURPOSE: To compare the computability of Observational Medical Outcomes Partnership (OMOP)-based queries related to prescreening of patients using two versions of the OMOP common data model (CDM; v5.3 and v5.4) and to assess the performance of the Greater Paris University Hospital (APHP) prescreening tool. MATERIALS AND METHODS: We identified the prescreening information items being relevant for prescreening of patients with cancer. We randomly selected 15 academic and industry-sponsored urology phase I-IV clinical trials (CTs) launched at APHP between 2016 and 2021. The computability of the related prescreening criteria (PC) was defined by their translation rate in OMOP-compliant queries and by their execution rate on the APHP clinical data warehouse (CDW) containing data of 205,977 patients with cancer. The overall performance of the prescreening tool was assessed by the rate of true- and false-positive cases of three randomly selected CTs. RESULTS: We defined a list of 15 minimal information items being relevant for patients' prescreening. We identified 83 PC of the 534 eligibility criteria from the 15 CTs. We translated 33 and 62 PC in queries on the basis of OMOP CDM v5.3 and v5.4, respectively (translation rates of 40% and 75%, respectively). Of the 33 PC translated in the v5.3 of the OMOP CDM, 19 could be executed on the APHP CDW (execution rate of 58%). Of 83 PC, the computability rate on the APHP CDW reached 23%. On the basis of three CTs, we identified 17, 32, and 63 patients as being potentially eligible for inclusion in those CTs, resulting in positive predictive values of 53%, 41%, and 21%, respectively. CONCLUSION: We showed that PC could be formalized according to the OMOP CDM and that the oncology extension increased their translation rate through better representation of cancer natural history.

Treatment patterns and outcomes among patients with small-cell lung cancer (SCLC) in Europe: a retrospective cohort study.

OBJECTIVE: Describe characteristics, treatment patterns and clinical outcomes of patients with small-cell lung cancer (SCLC). DESIGN: Retrospective chart review study defining several cohorts: (1) limited-stage disease (LD) SCLC initiating 1L therapy (1 L LD-SCLC), (2) extensive-stage disease (ED) SCLC initiating 1L therapy (1L ED-SCLC) and (3) patients initiating 2L therapy. SETTING: 39 physicians (medical oncologists, thoracic oncologists and/or pulmonologists) from France, Italy and the UK. PARTICIPANTS: Patients >18 years of age with a confirmed diagnosis of LD-SCLC or ED-SCLC and a full oncology medical history. Patients included initiated a 1L (2013-2015) or 2L (2013-2016) treatment (chemotherapy and/or radiotherapy-RT). PRIMARY AND SECONDARY OUTCOME MEASURES: Overall survival (OS) and progression-free survival (PFS). RESULTS: 231 patients in 1L LD-SCLC, 308 in 1L ED-SCLC and 225 with relapse/refractory SCLC initiating 2L treatment were included. The proportion of men was higher across all groups (56.8% to 68.5%) and mean age at time of diagnosis was 66.0 and 65.4 years in 1L LD-SCLC and 2L ED-SCLC cohorts. The majority of patients in LD-SCLC 1L group received chemotherapy with RT (76.2%). Patients initiating 2L therapy predominantly received chemotherapy alone (79.6%).Median OS in 1 L patients was 17.3 months in LD-SCLC and 8.8 months in ED-SCLC. Median PFS was 11.6 months in LD-SCLC and 6.1 months in ED-SCLC patients. Median OS in patients initiating 2L treatment was 6.6 months. OS from start of 2L treatment was lower in patients initially diagnosed with ED (5.1 months) than in patients initially diagnosed with LD (9.3 months) (p<0.0001). OS and PFS were assessed from the start of 1L or 2L therapy, depending on the cohort. CONCLUSIONS: Despite the availability of a high number of treatments and combinations, the prognosis of SCLC is still unsatisfactory, especially for those patients diagnosed with ED-SCLC, indicating high unmet need in this patient population.

Federated learning for predicting histological response to neoadjuvant chemotherapy in triple-negative breast cancer.

Triple-negative breast cancer (TNBC) is a rare cancer, characterized by high metastatic potential and poor prognosis, and has limited treatment options. The current standard of care in nonmetastatic settings is neoadjuvant chemotherapy (NACT), but treatment efficacy varies substantially across patients. This heterogeneity is still poorly understood, partly due to the paucity of curated TNBC data. Here we investigate the use of machine learning (ML) leveraging whole-slide images and clinical information to predict, at diagnosis, the histological response to NACT for early TNBC women patients. To overcome the biases of small-scale studies while respecting data privacy, we conducted a multicentric TNBC study using federated learning, in which patient data remain secured behind hospitals' firewalls. We show that local ML models relying on whole-slide images can predict response to NACT but that collaborative training of ML models further improves performance, on par with the best current approaches in which ML models are trained using time-consuming expert annotations. Our ML model is interpretable and is sensitive to specific histological patterns. This proof of concept study, in which federated learning is applied to real-world datasets, paves the way for future biomarker discovery using unprecedentedly large datasets.

Identification of Tissue of Origin and Guided Therapeutic Applications in Cancers of Unknown Primary Using Deep Learning and RNA Sequencing (TransCUPtomics).

Cancers of unknown primary (CUP) are metastatic cancers for which the primary tumor is not found despite thorough diagnostic investigations. Multiple molecular assays have been proposed to identify the tissue of origin (TOO) and inform clinical care; however, none has been able to combine accuracy, interpretability, and easy access for routine use. We developed a classifier tool based on the training of a variational autoencoder to predict tissue of origin based on RNA-sequencing data. We used as training data 20,918 samples corresponding to 94 different categories, including 39 cancer types and 55 normal tissues. The TransCUPtomics classifier was applied to a retrospective cohort of 37 CUP patients and 11 prospective patients. TransCUPtomics exhibited an overall accuracy of 96% on reference data for TOO prediction. The TOO could be identified in 38 (79%) of 48 CUP patients. Eight of 11 prospective CUP patients (73%) could receive first-line therapy guided by TransCUPtomics prediction, with responses observed in most patients. The variational autoencoder added further utility by enabling prediction interpretability, and diagnostic predictions could be matched to detection of gene fusions and expressed variants. TransCUPtomics confidently predicted TOO for CUP and enabled tailored treatments leading to significant clinical responses. The interpretability of our approach is a powerful addition to improve the management of CUP patients.

Impact of COVID-19 on the management of patients with thoracic cancers in a tertiary referral center.

INTRODUCTION: Severe acute respiratory syndrome coronavirus (SARS-CoV)-2 has spread worldwide in 2020 leading the World Health Organization to declare a pandemic. Patients with thoracic cancers have been reported at higher risk to develop severe disease, and die from COVID-19. In this setting, clinical practice recommendations for the management of patients were published. We report here how these guidelines were implemented in a routine practice setting. METHODS: We retrospectively collected the characteristics, treatment regimen and modification, as well as COVID-19 status and death for all patients with thoracic malignancies scheduled for an appointment at Institute Curie from March 23rd to April 17th 2020. RESULTS: A total of 339 patients were included. Treatment strategy was modified for a total of 110 (32 %) patients because of COVID-19; these modifications were in accordance with guidelines for 92 % of patients. The majority of dose modifications were related to immune checkpoint inhibitors, for which switch to flat dosing every 4-6 weeks was made. A total of 5 (1.5 %) patients were diagnosed with COVID-19 disease, 1 of whom died from disease complication. CONCLUSION: Our study provides a unique insight in the decision making for patients with thoracic malignancies in the setting of COVID-19 outbreak, showing how guidelines were implemented in the clinic, and what may be optimized in the clinical practice of thoracic oncology in the future.

High in vitro and in vivo synergistic activity between mTORC1 and PLK1 inhibition in adenocarcinoma NSCLC.

Significant rational is available for specific targeting of PI3K/AKT/mTOR pathway in the treatment of non-small cell lung cancer (NSCLC). However, almost all clinical trials that have evaluated Pi3K pathway-based monotherapies/combinations did not observe an improvement of patient's outcome. The aim of our study was therefore to define combination of treatment based on the determination of predictive markers of resistance to the mTORC1 inhibitor RAD001/Everolimus. An in vivo study showed high efficacy of RAD001 in NSCLC Patient-Derived Xenografts (PDXs). When looking at biomarkers of resistance by RT-PCR study, three genes were found to be highly expressed in resistant tumors, i.e., PLK1, CXCR4, and AXL. We have then focused our study on the combination of RAD001 + Volasertib, a PLK1 inhibitor, and observed a high antitumor activity of the combination in comparison to each monotherapy; similarly, a clear synergistic effect between the two compounds was found in an in vitro study. Pharmacodynamics study demonstrated that this synergy was due to (1) tumor vascularization decrease, increase of the HIF1 protein expression and decrease of the intracellular pH, and (2) decrease of the Carbonic Anhydrase 9 (CAIX) protein that could not correct intracellular acidosis. In conclusion, all these preclinical data strongly suggest that the inhibition of mTORC1 and PLK1 proteins may be a promising therapeutic approach for NSCLC patients.

Lymphovascular invasion after neoadjuvant chemotherapy is strongly associated with poor prognosis in breast carcinoma.

PURPOSE: Few studies evaluated the prognostic value of the presence of lymphovascular invasion (LVI) after neoadjuvant chemotherapy (NAC) for breast cancer (BC). METHODS: The association between LVI and survival was evaluated in a cohort of BC patients treated by NAC between 2002 and 2011. Five post-NAC prognostic scores (ypAJCC, RCB, CPS, CPS + EG and Neo-Bioscore) were evaluated and compared with or without the addition of LVI. RESULTS: Out of 1033 tumors, LVI was present on surgical specimens in 29.2% and absent in 70.8% of the cases. Post-NAC LVI was associated with impaired disease-free survival (DFS) (HR 2.54; 95% CI 1.96-3.31; P < 0.001), and the magnitude of this effect depended on BC subtype (Pinteraction = 0.003), (luminal BC: HR 1.83; P = 0.003; triple negative BC: HR 3.73; P < 0.001; HER2-positive BC: HR 6.21; P < 0.001). Post-NAC LVI was an independent predictor of local relapse, distant metastasis, and overall survival; and increased the accuracy of all five post-NAC prognostic scoring systems. CONCLUSIONS: Post-NAC LVI is a strong independent prognostic factor that: (i) should be systematically reported in pathology reports; (ii) should be used as stratification factor after NAC to propose inclusion in second-line trials or adjuvant treatment; (iii) should be included in post-NAC scoring systems.

Percutaneous vertebroplasty in vertebral metastases from breast cancer: interest in terms of pain relief and quality of life.

Percutaneous vertebroplasty (PV) is a therapeutic option in patients with vertebral metastases (VM). However its efficacy in pain relief, improvement in quality of life and safety in patients with VM from breast cancer has not been reported. We present a longitudinal retrospective study of 31 consecutively treated female patients with VM from breast cancer where 88 vertebrae were treated in 44 sessions of PV, in which osteolytic, osteoblastic and mixed lesions were recorded. The visual analogue pain scale (VAS) was used to evaluate pain pre-PV, at one, three, six and 12 months post-PV. The Eastern Cooperative Group (ECOG) performance status scale was used at the same time intervals to measure quality of life: 90.3% pain relief was identified with a VAS reduction from 5.7 ± 2.0 pre-PV to 2.9 ± 2.2 post-PV at one-month follow-up (p<0.001) and 0.6 ± 1.0 at 12-month follow-up (p<0.001). In our series 48.4% of patients were classified as having an ECOG grade 0 and 1 pre-PV, which increased to 80.8% at the 12-month follow-up. While 22.6% of the patients were classified at ECOG grades 3 and 4 pre-PV, this improved to 0% at 12 months follow-up. The morbidity rate for this procedure was 12.9% immediately and only 3.2% at 30 days post-PV with all complications being resolved medically or with CT-guided infiltration. PV is a safe procedure with a high efficacy in pain relief, and improvement of quality of life in patients with diverse types of VM from breast cancer.

Perception of care and experience of examination in women at risk of breast cancer undergoing intensive surveillance by standard imaging with or without MRI.

OBJECTIVE: Intensive surveillance in women at breast cancer risk is currently investigated in a French prospective, non-randomized, multicenter study, in which standard imaging--mammography±ultrasound ('Mx') and standard imaging combined with magnetic resonance imaging ('MRI') are compared with regard to perception of care and examination experience. METHODS: 1561 women were invited to complete the STAI-State Anxiety Inventory and breast cancer risk perception items at baseline (T0), and MGQ (MammoGraphy Questionnaire) and MRI discomfort items within 2 days after examinations (T1). RESULTS: Baseline compliance was high (>91%). Women from the 'MRI' group were significantly younger and displayed higher education level and risk perception. MRI discomfort related to the duration, immobility, prone position or noise was experienced by more than 20% of women. In multivariate analyses, 'MRI' was associated with more favorable examination psychological experience (p≤.001), especially in women younger than 50; baseline STAI-State anxiety was associated with lower MGQ scores (p≤.001) and higher MRI discomfort (p≤.001). CONCLUSION: In spite of the discomfort experienced with MRI, perception of care and experience with this surveillance procedure was more positive than with standard imaging. PRACTICE IMPLICATIONS: Information and support may assuage some of the adverse effects of an uncomfortable examination technique.

Anxiety and specific distress in women at intermediate and high risk of breast cancer before and after surveillance by magnetic resonance imaging and mammography versus standard mammography.

BACKGROUND: Intensive surveillance in women at intermediate and high breast cancer risk is currently investigated in a French prospective, non-randomized, multicentre study. Two surveillance modalities, standard imaging-mammography ± ultrasound ('Mx')-or standard imaging with magnetic resonance imaging ('MRI'), provided according to the level of breast cancer risk, are compared on psychological distress. METHODS: A total of 1561 women were invited to complete the State-Trait Anxiety Inventory (STAI), Impact of Event Scale (IES) Intrusion and Avoidance subscales and breast cancer-risk perception items at T0 (before examination) and T2 (1 to 3 months later) and the STAI-State anxiety at T1 (just after examination). Multiple regression analyses were performed. RESULTS: Baseline compliance was high (>91%). Between surveillance modalities, women differed significantly for age, education level, breast cancer-risk objective estimates and subjective perception. Mean STAI-State anxiety scores reflected low to moderate distress in both surveillance modalities. At baseline, MRI was associated with lower STAI-State anxiety (p ≤ 0.001) and Avoidance scores (p = 0.02), but at T1 and T2, no difference between surveillance modalities was observed on psychological outcomes. Abnormal surveillance result was associated with a higher STAI-State anxiety (p ≤ 0.01) and IES-Intrusion (p ≤ 0.01) scores; a personal history of breast cancer and higher risk perception was associated with higher psychological distress at T1 and T2. CONCLUSION: Standard breast imaging including MRI does not seem to convey more harmful psychological effects than standard imaging alone. Higher psychological distress observed in the case of history of breast cancer or higher breast cancer-risk perception evidences women with needs for specific support and information.

Use of a subcutaneous implantable pleural port in the management of recurrent malignant pleurisy: five-year experience based on 168 subcutaneous implantable pleural ports.

INTRODUCTION: The authors report their experience in the use of subcutaneous implantable pleural port (SIPP) catheters for the treatment of symptomatic recurrent malignant pleurisy. MATERIALS AND METHODS: Single-center, prospective follow-up of 137 patients (168 SIPPs). RESULTS: No SIPP placement failures were observed. All but 3 of the 125 evaluable patients obtained complete or partial relief of their dyspnea. Seventy-six patients (60.3%) were receiving chemotherapy. Spontaneous pleurodesis was observed within 2 months in 46 patients (36.8%). Twenty-six patients (20.8%) died during the month following SIPP placement. Forty-one patients (32%) survived for more than 6 months. The overall median survival time was 344 days. Three infectious complications (1 empyema, 2 cellulitis) and 3 mechanical complications were observed. The role of pleurodesis as prognostic factor was assessed. Seventy-one patients survived for more than 2 months, 36 with pleurodesis, 35 without pleurodesis, requiring repeated pleural aspiration. The difference observed between the two groups by the 120th day was no longer significant when chemotherapy was taken into account. CONCLUSION: SIPP is a safe and effective option for the outpatient management of recurrent malignant effusions and could be considered as first-line treatment in all patients with bilateral, compressive pleural effusion or poor lung reexpansion.

[Survival of breast cancers patients with meningeal carcinoma]. / Méningites carcinomateuses des cancers du sein.

PURPOSE: Among all solid tumors breast cancer is the most common cause of meningeal carcinomatosis (MC). The purpose of this study was to analyze clinical and biological responses as well as overall survival in MC patients (pts) of breast primary treated with intrathecal methotrexate (MTX). METHODS AND MATERIALS: Single-center retrospective series of MC pts treated between 2000 and 2007. Chemotherapy regimen was: MTX (15 mg/day; day 1-5) and depomedrol (40 mg, day 1) plus leucoverin (12 mg IV or 25 mg PO; day 1-5). Treatment cycles were repeated every 2 weeks. The survival was analyzed according to the characteristics of the tumor considering clinical and cytological response rates to treatment. RESULTS: The median survival was 4.5 months (range 0-53). In multivariate analysis, poor prognostic factors at diagnosis were: Performans status greater than 2 [P = 0.006, RR = 0.33 (0.15-0.71)], more than three chemotherapy regimens before MC diagnosis [P = 0.03, RR = 0.40 (0.19-0.93)], negative hormone receptor status [P = 0.02, RR = 0.4 (0.19-0.90)] and high Cyfra-21-1 level [P = 0.048, RR = 0.09-0.99]. The clinical progression after one cycle and the biological response after two cycles were independently correlated with OS [P<0.001, RR = 0.09 (0.02­0.37) and P = 0.003, RR = 3.6 (1.5­8.5), respectively]. A prognostic score designed to define three groups of patients is proposed. CONCLUSION: Although prognosis of patients with MC is poor, 1-year overall survival rate is 25%. The proposed prognostic score may be helpful in decision but warrants further assessment and validation in prospective trials.

Clinical relevance of human cancer xenografts as a tool for preclinical assessment: example of in-vivo evaluation of topotecan-based chemotherapy in a panel of human small-cell lung cancer xenografts.

Prediction of human tumor response based on preclinical data could reduce the failure rates of subsequent new anticancer drugs clinical development. Human small-cell lung carcinomas (SCLC) are characterized by high initial sensitivity to chemotherapy but a low median survival time because of drug resistance. The aim of this study was to evaluate the therapeutic relevance of a panel of human SCLC xenografts established in our laboratory using one compromising drug in SCLC, topotecan (TPT). Six SCLC xenografts derived from six patients were used: three were sensitive to a combination of etoposide (VP16), cisplatin (CDDP), and ifosfamide (IFO), and three were resistant, as published earlier. Growth inhibition was greater than 84% for five xenografts at doses of 1-2 mg/kg/day. TPT was combined with IFO, etoposide (VP16), and CDDP. IFO improved the efficacy of TPT in three of the five xenografts and complete responses were obtained even with the less TPT-sensitive xenograft. VP16 increased the efficacy of two of four xenografts and complete responses were obtained. The combination of TPT and CDDP did not improve TPT responses for any of the xenografts tested. Semiquantitative reverse transcriptase-PCR of genes involved in drug response, such as topoisomerase I, topoisomerase IIalpha, multidrug resistance 1 (MDR1), multidrug resistance-associated protein (MRP), lung resistance-related protein (LRP), and glutathione S-transferase pi (GSTpi), did not explain the variability in drug sensitivity between SCLC xenografts. In conclusion, these preclinical data mirror those from published clinical studies suggesting that our panel of SCLC xenografts represents a useful tool for preclinical assessment of new treatments.

Preclinical assessment of cisplatin-based therapy versus docetaxel-based therapy on a panel of human non-small-cell lung cancer xenografts.

The success of treatment of advanced non-small-cell lung cancer (NSCLC) remains very poor. The aim of this study was, on a series of NSCLC xenografts, to compare the efficacy of standard cisplatin-based or docetaxel-based chemotherapy. Seven human xenografts were obtained from six patients (two xenografts were derived from primary or metastatic tumors of the same patient). Three xenografts were adenocarcinomas and four were squamous cell carcinomas. All xenografts reproduced the same histology as that of the patient's original tumor. Docetaxel, administered as single-agent chemotherapy, induced a significant response in five of the seven NSCLC xenografts (71%), without significant increase after combination with cisplatin, vinorelbine, or gemcitabine. Relative expression of genes putatively involved in drug response was also studied in all xenografts and did not explain the variability of drug sensitivity. In conclusion, this panel of human NSCLC xenografts reliably reproduces the data obtained in patient tumors and the relative sensitivity to docetaxel reported in NSCLC patients.

Automatic acquisition of synonym resources and assessment of their impact on the enhanced search in EHRs.

OBJECTIVE: Currently, the use of natural language processing (NLP) approaches in order to improve search and exploration of electronic health records (EHRs) within healthcare information systems is not a common practice. One reason for this is the lack of suitable lexical resources. Indeed, in order to support such tasks, various types of such resources need to be collected or acquired (i.e., morphological, orthographic, synonymous). METHODS: We propose a novel method for the acquisition of synonymy resources. This method is language-independent and relies on existence of structured terminologies. It enables to decipher hidden synonymy relations between simple words and terms on the basis of their syntactic analysis and exploitation of their compositionality. RESULTS: Applied to series of synonym terms from the French subset of the UMLS , the method shows 99% precision. The overlap between thus inferred terms and the existing sparse resources of synonyms is very low. In order to better integrate these resources in an EHR search system, we analyzed a sample of clinical queries submitted by healthcare professionals. CONCLUSIONS: Observation of clinical queries shows that they make a very little use of the query expansion function, and, whenever they do, synonymy relations are rarely involved.

Automatic acquisition of synonyms from French UMLS for enhanced search of EHRs.

Currently, the use of Natural Language Processing (NLP) approaches in order to improve search and exploration of electronic health records (EHRs) within healthcare information systems is not a common practice. One reason for this is the lack of suitable lexical resources: various types of such resources need to be collected or acquired. In this work, we propose a novel method for the acquisition of synonymous resources. This method is language-independent and relies on existence of structured terminologies. It enables to decipher hidden synonymous relations between simple words and terms on the basis of their syntactic analysis and exploitation of their compositionality. Applied to series of synonym terms from the French subset of the UMLS, the method shows 99% precision. The overlap between thus inferred terms and the existing sparse resources of synonyms is very low.