A machine learning approach in a monocentric cohort for predicting primary refractory disease in Diffuse Large B-cell lymphoma patients.

INTRODUCTION: Primary refractory disease affects 30-40% of patients diagnosed with DLBCL and is a significant challenge in disease management due to its poor prognosis. Predicting refractory status could greatly inform treatment strategies, enabling early intervention. Various options are now available based on patient and disease characteristics. Supervised machine-learning techniques, which can predict outcomes in a medical context, appear highly suitable for this purpose. DESIGN: Retrospective monocentric cohort study. PATIENT POPULATION: Adult patients with a first diagnosis of DLBCL admitted to the hematology unit from 2017 to 2022. AIM: We evaluated in our Center five supervised machine-learning (ML) models as a tool for the prediction of primary refractory DLBCL. MAIN RESULTS: One hundred and thirty patients with Diffuse Large B-cell lymphoma (DLBCL) were included in this study between January 2017 and December 2022. The variables used for analysis included demographic characteristics, clinical condition, disease characteristics, first-line therapy and PET-CT scan realization after 2 cycles of treatment. We compared five supervised ML models: support vector machine (SVM), Random Forest Classifier (RFC), Logistic Regression (LR), Naïve Bayes (NB) Categorical classifier and eXtreme Gradient Boost (XGboost), to predict primary refractory disease. The performance of these models was evaluated using the area under the receiver operating characteristic curve (ROC-AUC), accuracy, false positive rate, sensitivity, and F1-score to identify the best model. After a median follow-up of 19.5 months, the overall survival rate was 60% in the cohort. The Overall Survival at 3 years was 58.5% (95%CI, 51-68.5) and the 3-years Progression Free Survival was 63% (95%CI, 54-71) using Kaplan-Meier method. Of the 124 patients who received a first line treatment, primary refractory disease occurred in 42 patients (33.8%) and 2 patients (1.6%) experienced relapse within 6 months. The univariate analysis on refractory disease status shows age (p = 0.009), Ann Arbor stage (p = 0.013), CMV infection (p = 0.012), comorbidity (p = 0.019), IPI score (p<0.001), first line of treatment (p<0.001), EBV infection (p = 0.008) and socio-economics status (p = 0.02) as influencing factors. The NB Categorical classifier emerged as the top-performing model, boasting a ROC-AUC of 0.81 (95% CI, 0.64-0.96), an accuracy of 83%, a F1-score of 0.82, and a low false positive rate at 10% on the validation set. The eXtreme Gradient Boost (XGboost) model and the Random Forest Classifier (RFC) followed with a ROC-AUC of 0.74 (95%CI, 0.52-0.93) and 0.67 (95%CI, 0.46-0.88) respectively, an accuracy of 78% and 72% respectively, a F1-score of 0.75 and 0.67 respectively, and a false positive rate of 10% for both. The other two models performed worse with ROC-AUC of 0.65 (95%CI, 0.40-0.87) and 0.45 (95%CI, 0.29-0.64) for SVM and LR respectively, an accuracy of 67% and 50% respectively, a f1-score of 0.64 and 0.43 respectively, and a false positive rate of 28% and 37% respectively. CONCLUSION: Machine learning algorithms, particularly the NB Categorical classifier, have the potential to improve the prediction of primary refractory disease in DLBCL patients, thereby providing a novel decision-making tool for managing this condition. To validate these results on a broader scale, multicenter studies are needed to confirm the results in larger cohorts.

[Marginal zone lymphoma associated with Reed-Sternberg cells: A challenge for the pathologist]. / Lymphome de la zone marginale ganglionnaire associé à des cellules de Reed-Sternberg : un challenge pour le pathologiste.

We report the case of a 46-year-old male patient presenting with a Claude Bernard-Horner Syndrome. Clinical evaluation showed a clonal B-cell population, lambda restricted. PET-scan captured femoral and axillary lymph nodes. Therefore the diagnosis of a marginal zone lymphoma was posted for which an attitude of watchful waiting was suggested. Eighteen months later, the patient developed an inguinal adenopathy. This lymph node led to the diagnosis of a nodular sclerosing Hodgkin lymphoma. Initial treatment with ABVD showed a good response, but the patient relapsed after eight months. A second biopsy confirmed the diagnosis of a marginal zone lymphoma but also identified giant Reed-Sternberg cells, (CD15+, CD30+ and CD20+). The initial biopsy was revised. This last diagnosis of a nodal marginal zone lymphoma with presence of Reed-Sternberg cells is rarely described in the literature. Several scientific theories can be found. Some cases described a transformation of non-Hodgkin lymphoma that presented Reed-Sternberg cells, other cases mentioned a collision or composite tumor. An accidental finding of Reed-Sternberg cells can be seen by after an infectious disease such as EBV. The presence of only Reed-Sternberg cells in a non-Hodgkin lymphoma is not sufficient to make a diagnosis of collision tumor.

Efficient SMN Rescue following Subcutaneous Tricyclo-DNA Antisense Oligonucleotide Treatment.

Spinal muscular atrophy (SMA) is a recessive disease caused by mutations in the SMN1 gene, which encodes the protein survival motor neuron (SMN), whose absence dramatically affects the survival of motor neurons. In humans, the severity of the disease is lessened by the presence of a gene copy, SMN2. SMN2 differs from SMN1 by a C-to-T transition in exon 7, which modifies pre-mRNA splicing and prevents successful SMN synthesis. Splice-switching approaches using antisense oligonucleotides (AONs) have already been shown to correct this SMN2 gene transition, providing a therapeutic avenue for SMA. However, AON administration to the CNS presents additional hurdles. In this study, we show that systemic delivery of tricyclo-DNA (tcDNA) AONs in a type III SMA mouse augments retention of exon 7 in SMN2 mRNA both in peripheral organs and the CNS. Mild type III SMA mice were selected as opposed to the severe type I model in order to test tcDNA efficacy and their ability to enter the CNS after maturation of the blood brain barrier (BBB). Furthermore, subcutaneous treatment significantly improved the necrosis phenotype and respiratory function. In summary, our data support that tcDNA oligomers effectively cross the blood-brain barrier and offer a promising systemic alternative for treating SMA.

Usefulness of a Gentle and Short Hemostasis Using the Transradial Band Device after Transradial Access for Percutaneous Coronary Angiography and Interventions to Reduce the Radial Artery Occlusion Rate (from the Prospective and Randomized CRASOC I, II, and III Studies).

The study sought to evaluate the benefit of the reduction in intensity and duration of the hemostasis obtained with the transradial (TR) Band compression device on the radial artery occlusion (RAO) rate. RAO is the most frequent complication of TR access for cardiac catheterization and limits future use of this safe route. Its occurrence must be minimized. Between 2009 and 2016, 3,616 TR accesses were randomized to TR Band hemostasis during 3 consecutive protocols: CRASOC I (Compression of Radial ArterieS without Occlusion): 13 versus 10 cc of air into the TR Band and for 4 hours of continuous compression; CRASOC II: 10 cc of air for 3 hours versus 2 hours of compression; and CRASOC III: 10 cc of air for 2 hours versus 1.5 hours of compression and virtual 4F introducer as default sheath (both arms). Radial artery patency was assessed by plethysmography at 24 hours, using Doppler for doubtful or negative plethysmography. The primary end point, 24 hours of RAO, was markedly reduced when hemostasis was soft (10 cc of air) and short (1.5 hours) and resulted in a 2.3% rate of RAO versus 9.4% for 13 cc, 4 hours. Hemostasis was obtained in 89% of patients with only 10 cc of air and in 97% of patients with less than the recommended 13 cc. About 8% of patients required more than the 1.5 hours of hemostasis time. In conclusion, short and soft hemostasis with the TR Band device leads to a low RAO rate.

Dasatinib and low-intensity chemotherapy in elderly patients with Philadelphia chromosome-positive ALL.

Prognosis of Philadelphia-positive (Ph(+)) acute lymphoblastic leukemia (ALL) in the elderly has improved during the imatinib era. We investigated dasatinib, another potent tyrosine kinase inhibitor, in combination with low-intensity chemotherapy. Patients older than age 55 years were included in the European Working Group on Adult ALL (EWALL) study number 01 for Ph(+) ALL (EWALL-PH-01 international study) and were treated with dasatinib 140 mg/day (100 mg/day over 70 years) with intrathecal chemotherapy, vincristine, and dexamethasone during induction. Patients in complete remission continued consolidation with dasatinib, sequentially with cytarabine, asparaginase, and methotrexate for 6 months. Maintenance therapy was dasatinib and vincristine/dexamethasone reinductions for 18 months followed by dasatinib until relapse or death. Seventy-one patients with a median age of 69 years were enrolled; 77% had a high comorbidity score. Complete remission rate was 96% and 65% of patients achieved a 3-log reduction in BCR-ABL1 transcript levels during consolidation. Only 7 patients underwent allogeneic hematopoietic stem cell transplantation. At 5 years, overall survival was 36% and up to 45% taking into account deaths unrelated to disease or treatment as competitors. Thirty-six patients relapsed, 24 were tested for mutation by Sanger sequencing, and 75% were T315I-positive. BCR-ABL1(T315I) was tested by allele-specific oligonucleotide reverse transcription-quantitative polymerase chain reaction in 43 patients and detection was associated with short-term relapses. Ten patients (23%) were positive before any therapy and 8 relapsed, all with this mutation. In conclusion, dasatinib combined with low-intensity chemotherapy was well-tolerated and gave long-term survival in 36% of elderly patients with Ph(+) ALL. Monitoring of BCR-ABL1(T315I) from diagnosis identified patients with at high risk of early relapse and may help to personalize therapy.

Gene and splicing therapies for neuromuscular diseases.

Neuromuscular disorders (NMD) are heterogeneous group of genetic diseases characterized by muscle weakness and wasting. Duchenne Muscular dystrophy (DMD) and Spinal muscular atrophy (SMA) are two of the most common and severe forms in humans and although the molecular mechanisms of these diseases have been extensively investigated, there is currently no effective treatment. However, new gene-based therapies have recently emerged with particular noted advances in using conventional gene replacement strategies and RNA-based technology. Whilst proof of principle have been demonstrated in animal models, several clinical trials have recently been undertaken to investigate the feasibility of these strategies in patients. In particular, antisense mediated exon skipping has shown encouraging results and hold promise for the treatment of dystrophic muscle. In this review, we summarize the recent progress of therapeutic approaches to neuromuscular diseases, with an emphasis on gene therapy and splicing modulation for DMD and SMA, focusing on the advantages offered by these technologies but also their challenges.

A rare recessive distal hereditary motor neuropathy with HSJ1 chaperone mutation.

OBJECTIVE: Distal hereditary motor neuropathies (dHMN) form a clinically and genetically heterogeneous group of disorders, characterized by muscle weakness and atrophy predominating at the distal part of the limbs, due to the progressive degeneration of motor neurons in the spinal cord. We report here a novel rare variant of dHMN with autosomal recessive inheritance in a large Jewish family originating from Morocco. The disease is characterized by a predominance of paralysis at the lower limbs and an early adulthood onset. We performed a genetic study in this family to identify and characterized the causing mutation. METHODS: Homozygosity mapping strategy and sequencing of the candidate genes were performed. Expression studies were made on patient fibroblasts. Functional experiments were performed on a cellular model of motor neuron disease. RESULTS: We mapped the disease to the 2q34-q36.1 chromosomal region and identified a homozygous splice mutation in the gene HSJ1 (DNAJB2) decreasing the expression of the 2 main isoforms HSJ1a and HSJ1b. Overexpression of both HSJ1a and HSJ1b reduced inclusion formation induced by the mutated SOD1-A4V in a neuronal cellular model. INTERPRETATION: HSJ1 is a neuronal enriched member of the HSP40/DNAJ co-chaperone family. Previous studies have shown that HSP40 proteins play a crucial role in protein aggregation and neurodegeneration in several neuronal types, in animal models and human diseases. Interestingly, this mutation causing a loss-of-function of HSJ1 is linked to a pure lower motor neuron disease, strongly suggesting that HSJ1 also plays an important and specific role in motor neurons.

Farnesyl transferase inhibitor (lonafarnib) in patients with myelodysplastic syndrome or secondary acute myeloid leukaemia: a phase II study.

Although an activating mutation of Ras is commonly observed in myelodysplastic syndrome (MDS), the role of Ras in the natural history of MDS remains largely unknown. We prospectively studied efficiency and tolerance of lonafarnib, a compound able to inhibit Ras signalling pathway through an inhibition of farnesyl transferase, in patients with MDS or secondary acute myeloid leukaemia (sAML). Lonafarnib was administered orally at a dose of 200 mg twice daily for three courses of 4 weeks (separated by 1 to 4 weeks without treatment). Sixteen patients were included: FAB/RAEB (n = 10), RAEB-T (n = 2), sAML (n = 2) and chronic myelomonocytic leukaemia (CMML; n = 2); WHO/RAEB-1 (n = 4), RAEB-2 (n = 5), AML (n = 5), CMML (n = 2). Median age was 70 (53-77) years. The karyotype was complex or intermediate in 11 patients, and the International Prognostic Scoring Systems (IPSS) risk groups were low in two patients, INT-1 in one patient, INT-2 in four patients and high in six patients (unknown or not applicable in three patients). Among the 14 patients tested, five had Ras mutations in codons 12, 13 or 61 of N-Ras, K-Ras or H-Ras. One patient was excluded of the analysis for protocol violation, and 15 patients were assessable for tolerance. Gastrointestinal toxicities (diarrhoea, nausea and anorexia) and myelosuppression were the major side effects. Other toxicities included infections, fatigue, increase of liver enzymes, arrhythmia and skin rash. One patient died of infection, and the treatment was stopped in one other who developed atrial fibrillation. Doses were reduced in all but one patient treated with more than one course of farnesyl transferase inhibitor. Responses were assessable in 12 patients. A partial response in one sAML patient and a very transient decrease of blast cell count with normalisation of karyotype in one MDS patient were observed. No relation between improvement of marrow parameters and detected Ras mutations was observed. Lonafarnib alone, administered following our schedule, has shown limited activity in patients with MDS or secondary AML. Gastrointestinal and haematological toxicities appear the limiting toxicity in this population of patients.