Multicenter automatic detection of invasive carcinoma on breast whole slide images.

Breast cancer is one of the most prevalent cancers worldwide and pathologists are closely involved in establishing a diagnosis. Tools to assist in making a diagnosis are required to manage the increasing workload. In this context, artificial intelligence (AI) and deep-learning based tools may be used in daily pathology practice. However, it is challenging to develop fast and reliable algorithms that can be trusted by practitioners, whatever the medical center. We describe a patch-based algorithm that incorporates a convolutional neural network to detect and locate invasive carcinoma on breast whole-slide images. The network was trained on a dataset extracted from a reference acquisition center. We then performed a calibration step based on transfer learning to maintain the performance when translating on a new target acquisition center by using a limited amount of additional training data. Performance was evaluated using classical binary measures (accuracy, recall, precision) for both centers (referred to as "test reference dataset" and "test target dataset") and at two levels: patch and slide level. At patch level, accuracy, recall, and precision of the model on the reference and target test sets were 92.1% and 96.3%, 95% and 87.8%, and 73.9% and 70.6%, respectively. At slide level, accuracy, recall, and precision were 97.6% and 92.0%, 90.9% and 100%, and 100% and 70.8% for test sets 1 and 2, respectively. The high performance of the algorithm at both centers shows that the calibration process is efficient. This is performed using limited training data from the new target acquisition center and requires that the model is trained beforehand on a large database from a reference center. This methodology allows the implementation of AI diagnostic tools to help in routine pathology practice.

Evaluation of Sample Pooling for SARS-CoV-2 Detection in Nasopharyngeal Swab and Saliva Samples with the Idylla SARS-CoV-2 Test.

Due to increased demand for testing, as well as restricted supply chain resources, testing for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection continues to face many hurdles. Pooling several samples has been proposed as an alternative approach to address these issues. We investigated the feasibility of pooling nasopharyngeal swab (NPS) or saliva samples for SARS-CoV-2 testing with a commercial assay (Idylla SARS-CoV-2 test; Biocartis). We evaluated the 10-pool and 20-pool approaches for 149 subjects, with 30 positive samples and 119 negative samples. The 10-pool approach had sensitivity of 78.95% (95% confidence interval [CI], 54.43% to 93.95%) and specificity of 100% (95% CI, 71.51% to 100%), whereas the 20-pool approach had sensitivity of 55.56% (95% CI, 21.20% to 86.30%) and specificity of 100% (95% CI, 25% to 100%). No significant difference was observed between the results obtained with pooled NPS and saliva samples. Given the rapidity, full automation, and practical advantages of the Idylla SARS-CoV-2 assay, pooling of 10 samples has the potential to significantly increase testing capacity for both NPS and saliva samples, with good sensitivity. IMPORTANCE To control outbreaks of coronavirus disease 2019 (COVID-19) and to avoid reagent shortages, testing strategies must be adapted and maintained for the foreseeable future. We analyzed the feasibility of pooling NPS and saliva samples for SARS-CoV-2 testing with the Idylla SARS-CoV-2 test, and we found that sensitivity was dependent on the pool size. The SARS-CoV-2 testing capacity with both NPS and saliva samples could be significantly expanded by pooling 10 samples; however, pooling 20 samples resulted in lower sensitivity.

Association of TRF2 expression and myeloid-derived suppressor cells infiltration with clinical outcome of patients with cutaneous melanoma.

The outcome of patients with cutaneous melanoma has been strongly modified by recent advances obtained with Immune Checkpoint Inhibitors (ICIs). However, despite this breakthrough, durable response to ICIs is limited to a subset of patients. We investigated whether the expression of TRF2, which preserves telomere integrity, and have an effect on tumor immunosurveillance notably by directly recruiting and activating myeloid-derived suppressor cells (MDSCs), could be a prognostic biomarker in patients with relapsed or metastatic melanoma based on different treatment regimens. We evaluated retrospectively the association of TRF2 expressed in melanoma cells in combination with intratumoral CD33+ CD15+ CD14- MDSCs, as detected by immunohistochemistry and quantified by digital analysis, to clinicopathological features and overall survival (OS) among 48 patients treated with ICIs and 77 patients treated with other treatment options. The densities/mm2 of TRF2+ cells (P=.003) and CD33+ cells (P=.004) were individually significantly related to poor OS. In addition, only the combined expression of CD33+/CD15+/CD14- cells/mm2 was significantly correlated to poor OS (P=.017) in the whole study population as well as in patients treated by ICIs (P=.023). There was no significant difference in OS when analyzing the other markers individually or in combination according to the treatment regimen. The pre-treatment assessment of TRF2 expression and CD33+ cells/mm2 along with the density of CD33+/CD15+/CD14- cells/mm2 could assess OS and better predict clinical response of patients with melanoma treated by ICIs.

Setting-Up a Rapid SARS-CoV-2 Genome Assessment by Next-Generation Sequencing in an Academic Hospital Center (LPCE, Louis Pasteur Hospital, Nice, France).

Introduction: Aside from the reverse transcription-PCR tests for the diagnosis of the COVID-19 in routine clinical care and population-scale screening, there is an urgent need to increase the number and the efficiency for full viral genome sequencing to detect the variants of SARS-CoV-2. SARS-CoV-2 variants assessment should be easily, rapidly, and routinely available in any academic hospital. Materials and Methods: SARS-CoV-2 full genome sequencing was performed retrospectively in a single laboratory (LPCE, Louis Pasteur Hospital, Nice, France) in 103 SARS-CoV-2 positive individuals. An automated workflow used the Ion Ampliseq SARS-CoV-2 panel on the Genexus Sequencer. The analyses were made from nasopharyngeal swab (NSP) (n = 64) and/or saliva (n = 39) samples. All samples were collected in the metropolitan area of the Nice city (France) from September 2020 to March 2021. Results: The mean turnaround time between RNA extraction and result reports was 30 h for each run of 15 samples. A strong correlation was noted for the results obtained between NSP and saliva paired samples, regardless of low viral load and high (>28) Ct values. After repeated sequencing runs, complete failure of obtaining a valid sequencing result was observed in 4% of samples. Besides the European strain (B.1.160), various variants were identified, including one variant of concern (B.1.1.7), and different variants under monitoring. Discussion: Our data highlight the current feasibility of developing the SARS-CoV-2 next-generation sequencing approach in a single hospital center. Moreover, these data showed that using the Ion Ampliseq SARS-CoV-2 Assay, the SARS-CoV-2 genome sequencing is rapid and efficient not only in NSP but also in saliva samples with a low viral load. The advantages and limitations of this setup are discussed.

Establishment of a Collection of Blood-Derived Products from COVID-19 Patients for Translational Research: Experience of the LPCE Biobank (Nice, France).

In only a few months after its inception, the COVID-19 pandemic lead to the death of hundreds of thousands of patients and to the infection of millions of people on most continents, mostly in the United States and in Europe. During this crisis, it was demonstrated that a better understanding of the pathogenicity, virulence, and contagiousness of SARS-CoV-2, all of which were initially underestimated, was urgently needed. The development of diagnostic tests to identify SARS-CoV-2 or to detect anti-SARS-CoV2 antibodies in blood, of vaccines, and of preventive and curative treatments has been relying on intense activity of scientists in academia and industry. It is noteworthy that these scientists depend on the use of high-quality biological samples taken from positive COVID-19 patients in a manner that preserves their integrity. Given this unique and emergent situation, it was necessary to urgently establish biological collections clinically annotated for immediate development of clinical and translational research projects focusing on COVID-19 biological aspects. It is in this very specific context that biobanks must rapidly adapt their infrastructure and/or operational capacity to fulfill new critical needs. We report the establishment of a biobank dedicated to the collection of blood-derived products (plasma, serum, and leukocytes) from COVID-19 patients hospitalized in the Nice Pasteur Hospital (Nice, France).

Real-world assessment of the BRAF status in non-squamous cell lung carcinoma using VE1 immunohistochemistry: A single laboratory experience (LPCE, Nice, France).

INTRODUCTION: International guidelines recommend BRAF mutational status assessment in treatment-naive advanced non-squamous non-small cell lung carcinoma (NSCLC) patients since the presence of a BRAFV600 mutation enables specific BRAF inhibitor treatment. For this purpose, the mutational status needs to be obtained in 10 working days. Herein, we prospectively evaluated the feasibility of systematic assessment of the BRAF status using immunohistochemistry (IHC) in a single institution (LPCE, Nice) at baseline for NSCLC diagnosed. METHODS: 1317 NSCLC were evaluated using BRAF IHC from 2011 to 2019. Initially the BRAF status was prospectively assessed using NGS and/or pyrosequencing in 618 consecutively diagnosed NSCLC patients from 2012 to 2016; BRAFV600E and BRAF nonV600E mutated tumors detected in this cohort were retrospectively evaluated using BRAF IHC. Secondarily, 699 biopsies of NSCLC were prospectively analyzed between 2017 and 2019 using BRAF IHC. BRAF IHC positive tumors were tested using a rapid BRAF specific PCR based assay. RESULTS: Initially, 21/618 (3%) of tumors (15 early and 6 late stage tumors) were BRAFV600E mutated according to the results of NGS and/or pyrosequencing. BRAF IHC was positive in 21/21 of these cases and negative in 51/51 (100 %) BRAF non V600E mutated cases. In the prospective BRAF IHC tested cohort of patients, 24/699 (3%) tumors (13 early and 11 late stage tumors) were positive with VE1 IHC. The BRAF PCR assay was positive in 20/24 (83 %) of these cases. CONCLUSION: BRAFV600E IHC screening of treatment-naïve NSCLC patients is a rapid, specific and very sensitive method which can lead in advanced stage positive NSCLC tumors to a BRAF inhibitor treatment. This test can be routinely integrated into mandatory predictive biomarker 'testing of NSCLC. According to the organization of patient care and the physician's request, this practice can be proposed as an alternative to NGS-based tissue biopsy made at baseline.

Targeted Assessment of the EGFR Status as Reflex Testing in Treatment-Naive Non-Squamous Cell Lung Carcinoma Patients: A Single Laboratory Experience (LPCE, Nice, France).

BACKGROUND: Assessment of actionable EGFR mutations is mandatory for treatment-naïve advanced or metastatic non-squamous lung carcinoma (NSLC), but the results need to be obtained in less than 10 working days. For rapid EGFR testing, an EGFR-specific polymerase chain reaction (PCR) assay is an alternative and simple approach compared to next generation sequencing (NGS). Here, we describe how a rapid EGFR-specific PCR assay can be implemented in a single laboratory center (LPCE, Nice, France) as reflex testing in treatment-naïve NSLC. METHODS: A total of 901 biopsies from NSLC with more than 10% of tumor cells were prospectively and consecutively evaluated for EGFR mutation status between November 2017 and December 2019 using the Idylla system (Biocartis NV, Mechelen, Belgium). NGS was performed for nonsmokers with NSLC wild type for EGFR, ALK, ROS1, and BRAF and with less than 50% PD-L1 positive cells using the Hotspot panel (Thermo Fisher Scientific, Waltham, MA, USA). RESULTS: Results were obtained from 889/901 (97%) biopsies with detection of EGFR mutations in 114/889 (13%) cases using the Idylla system. Among the 562 EGFR wild type tumors identified with Idylla, NGS detected one actionable and one nonactionable EGFR mutation. CONCLUSIONS: Rapid and targeted assessment of EGFR mutations in treatment-naïve NSLC can be implemented in routine clinical practice. However, it is mandatory to integrate this approach into a molecular algorithm that allows evaluation of potentially actionable genomic alterations other than EGFR mutations.