Ipsen is a French pharmaceutical company headquartered in Paris, France.[self-published source] It is specialized in oncology, neuroscience and rare diseases. It is publicly traded on the Euronext Paris as part of the SBF 120 index (2005).Ipsen, founded by Henri Beaufour in 1929, has 5,700 employees worldwide, and markets 20 drugs in 115 countries. The Beaufour family owns 57% of its shares and 73% of its voting rights,[self-published source] and two of its members, Anne Beaufour and Henri Beaufour, sit on its board of directors.
Lack of head-to-head comparative safety and efficacy data for abobotulinumtoxinA (aboBoNT-A) and onabotulinumtoxinA (onaBoNT-A) represented an important gap for informed spasticity management. This study compared the safety and efficacy of aboBoNT-A and onaBoNT-A for upper limb spasticity (ULS). DIRECTION (NCT04936542), a phase IV, randomized, double-blind, crossover study, involved 72 sites. Patients stratified by botulinum toxin A status (naïve/non-naïve) were randomized (1:1) to aboBoNT-A 900U followed by onaBoNT-A 360U (one cycle each), or vice versa. Muscles (wrist/finger flexors, biceps brachii) were injected with a fixed volume, using instrument-guided injection techniques. Participants fulfilling retreatment criteria received a second cycle at Week 12; otherwise, they were reassessed every 4 weeks (to Week 24) until requiring retreatment. Primary analyses tested non-inferiority based on treatment-emergent adverse events (TEAEs) from injection to Week 12 using a 5 Spasticity of the upper limb presents as muscle and limb stiffness that affects the arm and/or hand. It happens when parts of the brain or spinal cord that control movement are damaged. Spasticity can be painful and can make daily activities challenging. Some people with upper limb spasticity receive botulinum toxin A (BoNT-A) injections to help relax the affected muscles. BoNT-A injections work for a period, but retreatment is required when the effects start to wear off. AbobotulinumtoxinA and onabotulinumtoxinA are different types of BoNT-A. DIRECTION is the first-ever spasticity study to compare abobotulinumtoxinA and onabotulinumtoxinA directly in people with upper limb spasticity. It involved 464 people living with spasticity across the USA, France, and Canada. The study population had an average age of 57 years, two-thirds were male, and most had spasticity because of a stroke. The study found that, over the 12 weeks after BoNT-A treatment, similar numbers of patients had side-effects to abobotulinumtoxinA and onabotulinumtoxinA. In addition, it took longer for symptoms to return after treatment with abobotulinumtoxinA (99 days) compared with onabotulinumtoxinA (96 days). Longer BoNT-A treatment duration can reduce the number of days that people with upper limb spasticity have to live with symptoms before their next injection. More symptom-free days could reduce the need for patients and their caregivers to seek additional care and could reduce their overall treatment burden. These findings provide new information to help healthcare professionals make decisions about BoNT-A treatment to achieve the best outcomes for their patients with arm/hand spasticity.
Closed-source large language models (LLMs), such as ChatGPT and Gemini, are increasingly consulted for medical advice, yet their explanations may appear plausible while failing to reflect the model's underlying reasoning process. This gap poses serious risks as patients and clinicians may trust coherent but misleading explanations. We conduct a systematic black-box evaluation of faithfulness in medical reasoning among three widely used closed-source LLMs. Our study consists of three perturbation-based probes: (1) causal ablation, testing whether stated chain-of-thought (CoT) reasoning causally influences predictions; (2) positional bias, examining whether models create post-hoc justifications for answers driven by input positioning; and (3) hint injection, testing susceptibility to external suggestions. We complement these quantitative probes with a small-scale human evaluation of model responses to patient-style medical queries to examine concordance between physician assessments of explanation faithfulness and layperson perceptions of trustworthiness. We find that CoT reasoning steps often do not causally drive predictions, and models readily incorporate external hints without acknowledgment. In contrast, positional biases showed minimal impact in this setting. These results underscore that faithfulness, not just accuracy, must be central in evaluating LLMs for medicine, to ensure both public protection and safe clinical deployment.
BACKGROUND:For patients with advanced renal cell carcinoma, maintaining quality of life during treatment is a priority. The phase III CheckMate 9ER study demonstrated improved efficacy and health-related quality of life (HRQoL) outcomes with first-line cabozantinib plus nivolumab (CaboNivo) versus sunitinib (Sun) and underscored the need to improve understanding of the relationship between clinical efficacy and the patient treatment experience. In this work, we investigated the relationship between changes in HRQoL and clinical outcomes, as well as identifying symptom-based patient profiles that could predict disease course and clinical outcomes. MATERIALS AND METHODS:This post hoc analysis of CheckMate 9ER used the patient-reported outcomes 19-item Functional Assessment of Cancer Therapy-Kidney Symptom Index (FKSI-19) to evaluate HRQoL. Cox proportional modeling was used to investigate the association between early changes in FKSI-19 scores and clinical outcomes, while latent class analysis (LCA) was used to identify symptom-based patient profiles. RESULTS:The nature of patient-reported symptoms was similar at baseline and 13 weeks for CaboNivo and Sun, but symptom frequency varied between arms. Early worsening of bone pain [hazard ratio (HR) 1.45, P = 0.010] and sleep quality (HR 1.45, P = 0.007) were associated with increased risk of mortality. The LCA identified three distinct subgroups of patients by symptom burden and risk profile. Patients in the limited symptoms class at week 13 were more likely to have received CaboNivo (limited: 63.4%; moderate-to-severe: 45.9%; severe: 38.5%), had a longer duration of treatment (>122 weeks: 37.2% versus 14.7% versus 17.9%), and were generally less likely to experience disease progression (61.6% versus 55.1% versus 76.9%) than other symptom classes. CONCLUSIONS:This post hoc analysis identified an association between efficacy and HRQoL based on data from CheckMate 9ER. These results highlight the necessity of balancing treatment choice, tolerability, and HRQoL to optimize treatment outcomes and support shared decision-making.
