Therapeutic response in the HAWK and HARRIER trials using deep learning in retinal fluid volume and compartment analysis.

OBJECTIVES: To assess the therapeutic response to brolucizumab and aflibercept by deep learning/OCT-based analysis of macular fluid volumes in neovascular age-related macular degeneration. METHODS: In this post-hoc analysis of two phase III, randomised, multi-centre studies (HAWK/HARRIER), 1078 and 739 treatment-naive eyes receiving brolucizumab or aflibercept according to protocol-specified criteria in HAWK and HARRIER, respectively, were included. Macular fluid on 41,840 OCT scans was localised and quantified using a validated deep learning-based algorithm. Volumes of intraretinal fluid (IRF), subretinal fluid (SRF), pigment epithelial detachment (PED) for all central macular areas (1, 3 and 6 mm) in nanolitres (nL) and best corrected visual acuity (BCVA) change in ETDRS letters were associated using mixed models for repeated measures. RESULTS: Baseline IRF volumes decreased by >92% following the first intravitreal injection and consistently remained low during follow-up. Baseline SRF volumes decreased by >74% following the first injection, while PED volume resolved by 68-79% of its baseline volume. Resolution of SRF and PED was dependent on the substance and regimen used. Larger residual post-loading IRF, SRF and PED volumes were all independently associated with progressive vision loss during maintenance, where the differences in mean BCVA change between high and low fluid volume subgroups for IRF, SRF and PED were 3.4 letters (p < 0.0001), 1.7 letters (p < 0.001) and 2.5 letters (p < 0.0001), respectively. CONCLUSIONS: Deep-learning methods allow an accurate assessment of substance and regimen efficacy. Irrespectively, all fluid compartments were found to be important markers of disease activity and were relevant for visual outcomes.

Does real-time artificial intelligence-based visual pathology enhancement of three-dimensional optical coherence tomography scans optimise treatment decision in patients with nAMD? Rationale and design of the RAZORBILL study.

BACKGROUND/RATIONALE: Artificial intelligence (AI)-based clinical decision support tools, being developed across multiple fields in medicine, need to be evaluated for their impact on the treatment and outcomes of patients as well as optimisation of the clinical workflow. The RAZORBILL study will investigate the impact of advanced AI segmentation algorithms on the disease activity assessment in patients with neovascular age-related macular degeneration (nAMD) by enriching three-dimensional (3D) retinal optical coherence tomography (OCT) scans with automated fluid and layer quantification measurements. METHODS: RAZORBILL is an observational, multicentre, multinational, open-label study, comprising two phases: (a) clinical data collection (phase I): an observational study design, which enforces neither strict visit schedule nor mandated treatment regimen was chosen as an appropriate design to collect data in a real-world clinical setting to enable evaluation in phase II and (b) OCT enrichment analysis (phase II): de-identified 3D OCT scans will be evaluated for disease activity. Within this evaluation, investigators will review the scans once enriched with segmentation results (i.e., highlighted and quantified pathological fluid volumes) and once in its original (i.e., non-enriched) state. This review will be performed using an integrated crossover design, where investigators are used as their own controls allowing the analysis to account for differences in expertise and individual disease activity definitions. CONCLUSIONS: In order to apply novel AI tools to routine clinical care, their benefit as well as operational feasibility need to be carefully investigated. RAZORBILL will inform on the value of AI-based clinical decision support tools. It will clarify if these can be implemented in clinical treatment of patients with nAMD and whether it allows for optimisation of individualised treatment in routine clinical care.

A framework to guide dose & regimen strategy for clinical drug development.

Optimizing new drug therapies remains a challenge for clinical development, despite the use of ever more sophisticated quantitative methodologies. Although conceptually simple, the idea of finding the right treatment at the right dose for the right patient to ensure an appropriate balance of risks and benefits is challenging and requires a multidisciplinary approach. In this paper, we present a framework developed as a tool for organizing knowledge and facilitating collaboration in development teams.

Robust ΔΔct estimate.

The ΔΔct method estimates fold change in gene expression data from RT-PCR assay. The ΔΔct estimate aggregates replicates using mean and standard deviation (sd) and is not robust to outliers which are in practice often removed before the non-outlying replicates are aggregated. The alternative of using robust statistics such as median and median absolute deviation (MAD) to aggregate the replicates is not done in practice perhaps because the distribution of a robust ΔΔct estimate based on median and MAD is not straightforward to deduce. We introduce a robust ΔΔct estimate and deduce an approximate distribution for it. Simulations show that when data has outliers, the robust ΔΔct estimate compared to the non-robust ΔΔct estimate leads to significantly reduced confidence interval length and a coverage close to the nominal coverage. The analysis of an RT-PCR data from a Novartis clinical trial demonstrates benefit of a robust ΔΔct estimate.

A practical design for a dual-agent dose-escalation trial that incorporates pharmacokinetic data.

Traditionally, model-based dose-escalation trial designs recommend a dose for escalation based on an assumed dose-toxicity relationship. Pharmacokinetic data are often available but are currently only utilised by clinical teams in a subjective manner to aid decision making if the dose-toxicity model recommendation is felt to be too high. Formal incorporation of pharmacokinetic data in dose-escalation could therefore make the decision process more efficient and lead to an increase in the precision of the resulting recommended dose, as well as decreasing the subjectivity of its use. Such an approach is investigated in the dual-agent setting using a Bayesian design, where historical single-agent data are available to advise the use of pharmacokinetic data in the dual-agent setting. The dose-toxicity and dose-exposure relationships are modelled independently and the outputs combined in the escalation rules. Implementation of stopping rules highlight the practicality of the design. This is demonstrated through an example which is evaluated using simulation.

Effect of various diuretic treatments on rosiglitazone-induced fluid retention.

The efficacy of diuretics in the management of rosiglitazone (RSG)-induced fluid retention was evaluated in a multicenter, randomized, open-label, parallel-group, proof-of-concept study. Of 381 patients who had type 2 diabetes and were on treatment with sulfonylurea or sulfonylurea plus metformin, 260 (63% male, 37% female) showed evidence of volume expansion as defined by an absolute reduction in hematocrit (Hct) of > or =0.5% after 12 wk of rosiglitazone 4 mg twice daily. They were randomly assigned to five treatments for 7 d: (1) Continuation of RSG (RSG-C), (2) RSG + furosemide (RSG+FRUS), (3) RSG + hydrochlorothiazide (RSG+HCTZ), (4) RSG + spironolactone (RSG+SPIRO), and (5) discontinuation of RSG. The primary end point was change in Hct at day 7 of diuretic treatment phase, powered to compare each diuretic group and the RSG discontinuation with the control group of RSG-C, with adjustments for multiple testing. After 12 wk on RSG, Hct fell by mean of 2.92% (95% confidence interval [CI] -3.10 to -2.63%; P < 0.001) and extracellular fluid volume increased by 0.62 L/1.73 m(2) (95% CI 0.26 to 0.90 L/1.73 m(2); P < 0.001). After treatment, the RSG+SPIRO group only showed a mean increase in Hct of 0.24%. The estimated mean difference in Hct reduction was significant: 1.14% (95% CI 0.29 to 1.98%) for RSG+SPIRO (P = 0.004) and 0.87% (95% CI 0.03 to 1.71%) for RSG+HCTZ (P = 0.041) only. In additional analyses of between-diuretic treatment effects SPIRO induced a greater Hct rescue at 0.88% (95% CI -0.12 to 1.87%; P = 0.095) and extracellular fluid volume reduction of -0.75 L/1.73 m(2) (95% CI -1.52 to 0.03 L/1.73 m(2); P = 0.06) compared with FRUS, suggesting superiority in the management of RSG-associated fluid retention. There were no significant differences between SPIRO and HCTZ. These findings are consistent with peroxisome proliferator-activated receptor-gamma agonist activation of the epithelial sodium channel in the distal collecting duct, a site of action of SPIRO and a potential target for thiazide diuretics.