Solving the Evidence Interpretability Crisis in Health Technology Assessment: A Role for Mechanistic Models?

Health technology assessment (HTA) aims to be a systematic, transparent, unbiased synthesis of clinical efficacy, safety, and value of medical products (MPs) to help policymakers, payers, clinicians, and industry to make informed decisions. The evidence available for HTA has gaps-impeding timely prediction of the individual long-term effect in real clinical practice. Also, appraisal of an MP needs cross-stakeholder communication and engagement. Both aspects may benefit from extended use of modeling and simulation. Modeling is used in HTA for data-synthesis and health-economic projections. In parallel, regulatory consideration of model informed drug development (MIDD) has brought attention to mechanistic modeling techniques that could in fact be relevant for HTA. The ability to extrapolate and generate personalized predictions renders the mechanistic MIDD approaches suitable to support translation between clinical trial data into real-world evidence. In this perspective, we therefore discuss concrete examples of how mechanistic models could address HTA-related questions. We shed light on different stakeholder's contributions and needs in the appraisal phase and suggest how mechanistic modeling strategies and reporting can contribute to this effort. There are still barriers dissecting the HTA space and the clinical development space with regard to modeling: lack of an adapted model validation framework for decision-making process, inconsistent and unclear support by stakeholders, limited generalizable use cases, and absence of appropriate incentives. To address this challenge, we suggest to intensify the collaboration between competent authorities, drug developers and modelers with the aim to implement mechanistic models central in the evidence generation, synthesis, and appraisal of HTA so that the totality of mechanistic and clinical evidence can be leveraged by all relevant stakeholders.

Investigation of mixed model repeated measures analyses and non-linear random coefficient models in the context of long-term efficacy data.

The longitudinal data from 2 published clinical trials in adult subjects with upper limb spasticity (a randomized placebo-controlled study [NCT01313299] and its long-term open-label extension [NCT01313312]) were combined. Their study designs involved repeat intramuscular injections of abobotulinumtoxinA (Dysport®), and efficacy endpoints were collected accordingly. With the objective of characterizing the pattern of response across cycles, Mixed Model Repeated Measures analyses and Non-Linear Random Coefficient (NLRC) analyses were performed and their results compared. The Mixed Model Repeated Measures analyses, commonly used in the context of repeated measures with missing dependent data, did not involve any parametric shape for the curve of changes over time. Based on clinical expectations, the NLRC included a negative exponential function of the number of treatment cycles, with its asymptote and rate included as random coefficients in the model. Our analysis focused on 2 specific efficacy parameters reflecting complementary aspects of efficacy in the study population. A simulation study based on a similar study design was also performed to further assess the performance of each method under different patterns of response over time. This highlighted a gain of precision with the NLRC model, and most importantly the need for its assumptions to be verified to avoid potentially biased estimates. These analyses describe a typical situation and the conditions under which non-linear mixed modeling can provide additional insights on the behavior of efficacy parameters over time. Indeed, the resulting estimates from the negative exponential NLRC can help determine the expected maximal effect and the treatment duration required to reach it.

Efficacy and safety of abobotulinumtoxinA liquid formulation in cervical dystonia: A randomized-controlled trial.

BACKGROUND: Approved botulinum toxin A products require reconstitution. AbobotulinumtoxinA solution for injection is a ready-to-use liquid formulation of abobotulinumtoxinA. OBJECTIVES: The objective of this study was to demonstrate the superior efficacy of abobotulinumtoxinA solution for injection to placebo and to test the noninferior efficacy of abobotulinumtoxinA solution for injection versus abobotulinumtoxinA (dry formulation) in cervical dystonia. METHODS: This was a phase-3, multicenter, prospective, double-blind, randomized, active, and placebo-controlled study (N = 369). Patients with cervical dystonia were randomized (3:3:1) to abobotulinumtoxinA solution for injection 500 U, abobotulinumtoxinA 500 U, or placebo. Following the double-blind phase, patients received abobotulinumtoxinA solution for injection, open-label, for up to 4 cycles. The primary outcome was change from baseline at week 4 of the Toronto Western Spasmodic Torticollis Rating Scale total score. Secondary measures included change from baseline or cycle baseline in Toronto Western Spasmodic Torticollis Rating Scale scores. RESULTS: At week 4, both products were superior to placebo (Toronto Western Spasmodic Torticollis Rating Scale total score least square mean decrease from baseline, abobotulinumtoxinA solution for injection 500 U -12.5, abobotulinumtoxinA 500 U -14.0, placebo -3.9; P < .0001 vs placebo). The noninferiority limit of 3 points in the Toronto Western Spasmodic Torticollis Rating Scale total score at week 4 was not met for abobotulinumtoxinA solution for injection versus abobotulinumtoxinA. Toronto Western Spasmodic Torticollis Rating Scale total score reductions were maintained for up to 4 cycles of abobotulinumtoxinA solution for injection open-label follow-up treatment. Safety profiles of abobotulinumtoxinA solution for injection and abobotulinumtoxinA were similar, with dysphagia and injection-site pain the most frequent drug-related adverse events. CONCLUSIONS: Although the predefined noninferiority criterion was not met, abobotulinumtoxinA solution for injection was similarly effective to freeze-dried abobotulinumtoxinA in reducing Toronto Western Spasmodic Torticollis Rating Scale total scores with a similar safety profile. AbobotulinumtoxinA solution for injection efficacy was maintained with chronic open-label treatment, and this novel formulation may add convenience as well as dosing accuracy to treatment with abobotulinumtoxinA. © 2016 International Parkinson and Movement Disorder Society.

AbobotulinumtoxinA for Equinus Foot Deformity in Cerebral Palsy: A Randomized Controlled Trial.

BACKGROUND: Although botulinum toxin is a well-established treatment of focal spasticity in cerebral palsy, most trials have been small, and few have simultaneously assessed measures of muscle tone and clinical benefit. METHODS: Global, randomized, controlled study to assess the efficacy and safety of abobotulinumtoxinA versus placebo in cerebral palsy children with dynamic equinus foot deformity. Patients were randomized (1:1:1) to abobotulinumtoxinA 10 U/kg/leg, 15 U/kg/leg, or placebo injections into the gastrocnemius-soleus complex (1 or both legs injected). In the primary hierarchical analysis, demonstration of benefit for each dose required superiority to placebo on the primary (change in Modified Ashworth Scale from baseline to week 4) and first key secondary (Physician's Global Assessment at week 4) end points. RESULTS: Two hundred and forty-one patients were randomized, and 226 completed the study; the intention to treat population included 235 patients (98%). At week 4, Modified Ashworth Scale scores significantly improved with abobotulinumtoxinA; mean (95% confidence interval) treatment differences versus placebo were -0.49 (-0.75 to -0.23; P = .0002) for 15 U/kg/leg and -0.38 (-0.64 to -0.13; P = .003) for 10 U/kg/leg. The Physician's Global Assessment treatment differences versus placebo of 0.77 (0.45 to 1.10) for 15 U/kg/leg and 0.82 (0.50 to 1.14) for 10 U/kg/leg were also significant (both Ps < .0001). The most common treatment-related adverse event was muscular weakness (10 U/Kg/leg = 2; placebo = 1). CONCLUSIONS: AbobotulinumtoxinA improves muscle tone in children with dynamic equinus resulting in an improved overall clinical impression and is well tolerated.

Angiogenic gene therapy does not cause retinal pathology.

BACKGROUND: The potential negative influence of angiogenic gene therapy on the development or progression of retinal pathologies such as diabetic retinopathy (DR) or age-related macular degeneration (AMD) has led to the systematic exclusion of affected patients from trials. We investigated the role of nonviral fibroblast factor 1 (NV1FGF) in two phase II, multinational, double-blind, randomized, placebo-controlled, gene therapy trials (TALISMAN 201 and 211). METHODS: One hundred and fifty-two subjects with critical limb ischemia or claudication were randomized to receive eight intramuscular injections of 2.5 ml of NV1FGF at 0.2 mg/ml or 0.4 mg/dl or placebo. One hundred and fifty-two patients received a plasmid dose of NV1FGF of up to 32 mg or placebo. All patients underwent a systematic ophthalmologic examination at baseline and at 3, 6 or 12 months following gene therapy. Twenty-six of these patients (Münster subgroup) received a retinal fluorescence angiography at baseline and at final examination. RESULTS: Among those 26 patients, four of nine patients with diabetes suffered from nonproliferative DR. Three patients showed non-exsudative AMD. No change of retinal morphology or function was observed in Münster subgroup of both TALISMAN trials independent of the intramuscular NV1FGF dosage applied. CONCLUSIONS: Angiogenic gene therapy using NV1FGF is safe even in diabetics.

Characterizing short stature by insulin-like growth factor axis status and genetic associations: results from the prospective, cross-sectional, epidemiogenetic EPIGROW study.

CONTEXT: Serum IGF-I levels are often low in patients with short stature (SS) without defined etiology. Hence, genetic investigations have focused on the GH-IGF-I axis. OBJECTIVE: Our objectives were to characterize IGF-I axis status and search for a broader range of genetic associations in children with SS and normal GH. DESIGN AND SETTING: We conducted a prospective, cross-sectional, epidemiogenetic case-control study in 9 European countries (2008-2010). PARTICIPANTS: Children (n = 275) aged ≥2 years with SS without defined etiology (≤-2.5 height SD score [SDS]) and ≥1 peak GH ≥7 µg/L) were recruited. METHODS: Serum IGF-I, IGF-binding protein-3 (IGFBP-3), and acid-labile subunit (ALS) levels were measured in a central laboratory. Candidate gene exome sequencing was performed in this cohort and ethnicity-matched controls. RESULTS: Serum IGF-I, IGFBP-3, and ALS levels were highly correlated, but there was a discrepancy between prevalence of IGF-I, IGFBP-3, and ALS deficiencies (53%, 30%, and 0.8%, respectively). An insertion-deletion (Indel) on the IGF1 gene (P = 1.2 × 10(-5), Bonferroni-corrected; case vs control frequency: 0.04 vs 0.112), an Indel on NFKB1 (P = 1.36 × 10(-10); case vs control frequency: 0.464 vs 0.272), and 2 single-nucleotide polymorphisms on ZBTB38 (P < 2.3 × 10(-6)) were associated with SS. At P < 10(-4), single-nucleotide polymorphisms on genes related to protein kinase regulation, MAPK, and Fanconi pathways were also associated with SS. CONCLUSIONS: IGF-I deficiency is a common feature in SS without defined etiology; an Indel in the IGF1 gene was associated with SS. However, genes involved in transcriptional regulation (NFKB1 and ZBTB38) and growth factor signaling were also associated, providing further candidates for genetic investigations on individual patients.

Characterization of a large panel of patient-derived tumor xenografts representing the clinical heterogeneity of human colorectal cancer.

PURPOSE: Patient-derived xenograft models are considered to represent the heterogeneity of human cancers and advanced preclinical models. Our consortium joins efforts to extensively develop and characterize a new collection of patient-derived colorectal cancer (CRC) models. EXPERIMENTAL DESIGN: From the 85 unsupervised surgical colorectal samples collection, 54 tumors were successfully xenografted in immunodeficient mice and rats, representing 35 primary tumors, 5 peritoneal carcinoses and 14 metastases. Histologic and molecular characterization of patient tumors, first and late passages on mice includes the sequence of key genes involved in CRC (i.e., APC, KRAS, TP53), aCGH, and transcriptomic analysis. RESULTS: This comprehensive characterization shows that our collection recapitulates the clinical situation about the histopathology and molecular diversity of CRC. Moreover, patient tumors and corresponding models are clustering together allowing comparison studies between clinical and preclinical data. Hence, we conducted pharmacologic monotherapy studies with standard of care for CRC (5-fluorouracil, oxaliplatin, irinotecan, and cetuximab). Through this extensive in vivo analysis, we have shown the loss of human stroma cells after engraftment, observed a metastatic phenotype in some models, and finally compared the molecular profile with the drug sensitivity of each tumor model. Through an experimental cetuximab phase II trial, we confirmed the key role of KRAS mutation in cetuximab resistance. CONCLUSIONS: This new collection could bring benefit to evaluate novel targeted therapeutic strategies and to better understand the basis for sensitivity or resistance of tumors from individual patients.

Minimal effective dose of dysport and botox in a rat model of neurogenic detrusor overactivity.

BACKGROUND: Two botulinum toxins A have been evaluated for the treatment of refractory neurogenic detrusor overactivity (NDO) in humans: Dysport (abobotulinumtoxinA) and Botox (onabotulinumtoxinA). However, these two distinct commercialized products have different potency units and are not interchangeable. OBJECTIVE: Assessment of the dose response and determination of minimal effective dose (MED) for Dysport and Botox in spinal cord-injured (SCI) rats with NDO. DESIGN, SETTING, AND PARTICIPANTS: Female, adult, Sprague-Dawley rats (n=98) underwent T8-T9 spinal cord transection. Nineteen days after spinal cord injury, rats received intradetrusor injections (25µl injected, eight sites) of vehicle (V); Dysport 2, 5, 7.5, 10, and 12.5 U; and Botox 0.8, 2, 5, 7.5, and 10 U. Two days after injection, continuous cystometry was performed in conscious rats. MEASUREMENTS: Voiding contractions (VC) were assessed by duration of VC, intercontraction interval, voided volume, maximal pressure, pressure threshold change, and intravesical baseline pressure (BP), while nonvoiding contractions (NVC) were evaluated by amplitude, frequency, and volume threshold to elicit NVC. MEDs for Dysport and Botox were determined by analysis of variance step-down trend test. RESULTS AND LIMITATIONS: MEDs for Dysport and Botox were 10 U and 7.5 U, respectively. Regarding VC, only BP significantly decreased after 10 U Dysport and 7.5 U Botox compared to V (from 3.7±0.6 to 1.5±0.1 and 1.4±0.3mm Hg, respectively; p<0.01 and p<0.001, respectively). Dysport (10 U) and Botox (7.5 U) significantly inhibited NVC by decreasing their amplitude (from 7.4±1.1 to 5.8±0.5 and 5.4±0.6mm Hg, respectively; p<0.05); frequency (from 2.2±0.4 to 1.5±0.2 and 1.3±0.3 NVC per minute, respectively; p<0.01); and increasing volume threshold to elicit NVC (from 29.8±3.7 to 47.6±6.9 and 47.7±6.3%, respectively; p<0.05 and p<0.001, respectively). CONCLUSIONS: This is the first preclinical dose-ranging study with Dysport and Botox under standardized conditions showing similar inhibiting effects on NDO, albeit at different MEDs. It highlights the importance of distinguishing each preparation for predicted outcomes and doses to be used. Further studies in patients with NDO are warranted to confirm these experimental results.

Therapeutic angiogenesis with intramuscular NV1FGF improves amputation-free survival in patients with critical limb ischemia.

This study evaluated the efficacy and safety of intramuscular administration of NV1FGF, a plasmid-based angiogenic gene delivery system for local expression of fibroblast growth factor 1 (FGF-1), versus placebo, in patients with critical limb ischemia (CLI). In a double-blind, randomized, placebo-controlled, European, multinational study, 125 patients in whom revascularization was not considered to be a suitable option, presenting with nonhealing ulcer(s), were randomized to receive eight intramuscular injections of placebo or 2.5 ml of NV1FGF at 0.2 mg/ml on days 1, 15, 30, and 45 (total 16 mg: 4 x 4 mg). The primary end point was occurrence of complete healing of at least one ulcer in the treated limb at week 25. Secondary end points included ankle brachial index (ABI), amputation, and death. There were 107 patients eligible for evaluation. Improvements in ulcer healing were similar for use of NV1FGF (19.6%) and placebo (14.3%; P = 0.514). However, the use of NV1FGF significantly reduced (by twofold) the risk of all amputations [hazard ratio (HR) 0.498; P = 0.015] and major amputations (HR 0.371; P = 0.015). Furthermore, there was a trend for reduced risk of death with the use of NV1FGF (HR 0.460; P = 0.105). The adverse event incidence was high, and similar between the groups. In patients with CLI, plasmid-based NV1FGF gene transfer was well tolerated, and resulted in a significantly reduced risk of major amputation when compared with placebo.

Therapeutic Angiogenesis With Intramuscular NV1FGF Improves Amputation-free Survival in Patients With Critical Limb Ischemia.

This study evaluated the efficacy and safety of intramuscular administration of NV1FGF, a plasmid-based angiogenic gene delivery system for local expression of fibroblast growth factor 1 (FGF-1), versus placebo, in patients with critical limb ischemia (CLI). In a double-blind, randomized, placebo-controlled, European, multinational study, 125 patients in whom revascularization was not considered to be a suitable option, presenting with nonhealing ulcer(s), were randomized to receive eight intramuscular injections of placebo or 2.5 ml of NV1FGF at 0.2 mg/ml on days 1, 15, 30, and 45 (total 16 mg: 4 × 4 mg). The primary end point was occurrence of complete healing of at least one ulcer in the treated limb at week 25. Secondary end points included ankle brachial index (ABI), amputation, and death. There were 107 patients eligible for evaluation. Improvements in ulcer healing were similar for use of NV1FGF (19.6%) and placebo (14.3%; P = 0.514). However, the use of NV1FGF significantly reduced (by twofold) the risk of all amputations [hazard ratio (HR) 0.498; P = 0.015] and major amputations (HR 0.371; P = 0.015). Furthermore, there was a trend for reduced risk of death with the use of NV1FGF (HR 0.460; P = 0.105). The adverse event incidence was high, and similar between the groups. In patients with CLI, plasmid-based NV1FGF gene transfer was well tolerated, and resulted in a significantly reduced risk of major amputation when compared with placebo.

Does the prognosis of cardiac arrest differ in trauma patients?

OBJECTIVE: It is proposed to not resuscitate trauma patients who have a cardiac arrest outside the hospital because they are assumed to have a dismal prognosis. Our aim was to compare the outcome of patients with traumatic or nontraumatic ("medical") out-of-hospital cardiac arrest. DESIGN: Cohort analysis of patients with out-of-hospital cardiac arrest included in the European Epinephrine Study Group's trial comparing high vs. standard doses of epinephrine. SETTING: Nine French university hospitals. PATIENTS: A total of 2,910 patients. INTERVENTIONS: Patients were successively and randomly assigned to receive repeated high doses (5 mg each) or standard doses (1 mg each) of epinephrine at 3-min intervals. MEASUREMENTS AND MAIN RESULTS: Return of spontaneous circulation, survival to hospital admission and discharge, and secondary outcome measures of 1-yr survival and neurologic outcome were recorded. In the trauma group, patients were younger (42 +/- 17 vs. 62 +/- 17 yrs, p < .001), presented with fewer witnessed out-of-hospital cardiac arrests (62.3% vs. 79.7%), and had fewer instances of ventricular fibrillation as the first documented pulseless rhythm (3.4% [95% confidence interval, 1.2-5.5%] vs. 17.3% [15.8-18.7%]). A return of spontaneous circulation was observed in 91 of 268 trauma patients (34.0% [28.3-39.6%]) compared with 797 of 2,642 medical patients (30.2% [28.4-31.9%]), and more trauma patients survived to be admitted to the hospital (29.9% [24.4-35.3%] vs. 23.5% [22.0-25.2%]). However, there was no significant difference between trauma and medical groups at hospital discharge (2.2% [0.5-4.0%] vs. 2.8% [2.1-3.4%]) and 1-yr survival (1.9% [0.3-3.5%] vs. 2.5% [1.9-3.1%]). Among patients who were discharged, a good neurologic status was observed in two trauma patients (33.3% [4.3-77.7%]) and 37 medical patients (50% [38.1-61.9%]). CONCLUSIONS: The survival and neurologic outcome of out-of-hospital cardiac arrest were not different between trauma and medical patients. This result suggests that, under the supervision of senior physicians, active resuscitation after out-of-hospital cardiac arrest is as important in trauma as in medical patients.