Apomorphine for prolonged disorders of consciousness: a multimodal open-label study

Abstract

Background: Apomorphine is a dopaminergic candidate therapy to improve recovery in patients with prolonged disorders of consciousness (PDoC). Behavioural improvements were previously described in non-controlled case series, but its efficacy and neural mechanisms remain largely unknown. This open-label controlled study using multimodal outcome measures investigates the action of apomorphine in severely brain-injured patients.

Methods: Thirteen PDoC patients received 30-day subcutaneous apomorphine treatment (n = 6) or standard care (control group, n = 7) in a neurological rehabilitation centre between February 2018 and January 2021. The apomorphine group was monitored 30 days before treatment initiation, during treatment and one year after treatment. Primary outcome measure was defined as changes in behavioural diagnosis using the Coma Recovery Scale-Revised (CRS-R). CRS-R index, recovery of new conscious behaviours, DoC-feeling scores, high-density electroencephalography, and fluorodeoxyglucose positron emission tomography were employed as secondary outcome measures. The control group was monitored with repeated CRS-R only. Registration: EudraCT 2018-003144-23; Clinicaltrials.govNCT03623828.

Findings: Groups (apomorphine vs. control: odds ratio 8.9, 95% CI 3.3-17.8) and study phase (treatment vs. baseline, apomorphine group only: odds ratio 3.9, 95% CI 1.5-10.1) significantly influenced positive changes in behavioural diagnosis. At one-year post-injury, 4/6 patients in the apomorphine group and 1/7 patients in the control group had improved their diagnosis. Similarly, CRS-R index was significantly influenced by study phase (treatment vs. baseline). All items on the DoC-feeling score were rated higher after treatment than before by both family and medical staff. Patients in the apomorphine group recovered more conscious behaviours than control patients. Alpha-band whole-brain connectivity and participation coefficient, as well as alpha-band parieto-temporal connectivity and frontal participation coefficient were higher after treatment than at baseline. Whole-brain metabolism increased by a relative mean of 13.8% after treatment compared to baseline, with a significant effect of timing (pre-vs. post-treatment scans) on regional SUV.

Interpretation: Long-lasting consciousness improvements were observed in patients treated with apomorphine, compared to controls and compared to baseline. Changes in brain connectivity and metabolism were observed after treatment, providing insights into possible neurophysiological mechanisms and target areas. This open-label study confirmed the feasibility and safety of apomorphine treatment, which may represent a key therapeutic option for PDoC.

Funding: University and University Hospital of Liege, Belgian National Funds for Scientific Research, Fund Generet of the King Baudouin Foundation, AstraZeneca Foundation, Leon Fredericq Foundation and NeuroHealing Pharmaceuticals Inc.

Keywords: Apomorphine; Coma; Disorders of consciousness; High-density electroencephalography; Minimally conscious state; Neuroimaging; Positron emission tomography; Severe brain injury; Treatment; Unresponsive wakefulness syndrome; Vegetative state.

Fig. 1

Study flow-chart. Adapted from the CONSORT 2010 Flow Diagram. tDCS, transcranial direct current stimulation. All 48 patients admitted to the rehabilitation centre within the study period were screened for eligibility. Fifteen patients were deemed eligible, of which eight were included in the apomorphine group and seven in the control group as they could not benefit from active treatment for logistical reasons or enrolment in the control group of other studies. Note that the ineligibility rate (33/48) observed in this study is highly specific to the recruiting rehabilitation centre as it admits patients for short rehabilitation stays, with locked-in syndrome or already in emergence from the minimally conscious state (accounting for 23/33 exclusions).

Fig. 2

Timeline of assessments. All patients were followed with the Coma Recovery Scale—Revised (CRS-R) and monitored for adverse events, while the apomorphine treatment group also underwent a multimodal assessment before and after the treatment phase. An optional treatment extension was proposed only to patients who showed clinical improvement during the treatment period. D/M, days/months after inclusion; hdEEG, high-density electroencephalography; FDG-PET, fluorodeoxyglucose positron emission tomography; DoC-feeling, extended version of the disorders of consciousness subjective feeling score.

Fig. 3

Longitudinal evolution of behavioural outcome measures. A. Evolution of behavioural diagnosis in the two groups across four time windows. For each patient, the most frequent diagnosis for all observations made within the given time window are displayed. B. Evolution of the CRS-R index (median—interquartile range) for the apomorphine (green) and the control (black) groups. For each patient, the median CRS-R index value was computed for all observations within each time window, and the median and quartile values were then calculated for each group. M, months; UWS, unresponsive wakefulness syndrome; MCS−, minimally conscious state minus; MCS+, minimally conscious state plus; EMCS, emergence from the MCS; CRS-R, Coma Recovery Scale–Revised.

Fig. 4

Effect sizes of the mixed-effects model on primary outcome (behavioural diagnosis). Forest plot displaying the decimal logarithm of the odds ratio to predict a change in behavioural diagnosis for the fixed effects included in the mixed-effects ordered logistic regression models, in the two analysis conditions (all patients and apomorphine group only). Apomorphine: odds ratio of the apomorphine group compared to the control group, Treatment and Washout phase: odds ratio compared to the baseline phase. Asterisks denote significant effects and whiskers indicate 95% confidence intervals. Interpretation of confounder coefficients is subject to special caution and may lead to misleading conclusions.

Fig. 5

Presence of new behaviours. The upper panel displays the percentage of patients in each group who demonstrated new behaviours during treatment and washout phases. The lower panel shows the number of new behaviours developed per patient in each group for the same period. A new behaviour is defined as a Coma Recovery Scale–Revised (CRS-R) item which was never previously reported during baseline assessments. Treatment and washout phases for the control group are defined as day 31–60 and 61–90 after inclusion, respectively (similar to the apomorphine group). CRS-R items (without arousal) were grouped into reflexive or conscious according to the scale guidelines, and conscious behaviours were further divided into language- and non-language-related.

Fig. 6

DoC-feeling questionnaire results. Ratings by family (one rater) and staff (average of three raters) pre- and post-treatment on an eight-item clinical questionnaire using visual analogue scale reports of perceived patient status over the last seven days. The post-treatment questionnaire featured two additional items on treatment efficacy and tolerance.

Fig. 7

High-density electroencephalography. Whole-brain connectivity (A) and participation coefficient (B) values in the alpha band are displayed for patients in the apomorphine group before and after treatment, as well as for 26 healthy controls. 3D topographs of alpha mean brain network connectivity and 2D topographs of alpha participation coefficient averaged over the six patients pre- and post-treatment, colour-coded for connection/participation strength (from blue to red), are displayed over the corresponding boxplots. Whole-brain connectivity and participation coefficient are increased in the post-treatment condition compared to before treatment for the alpha frequency band, returning to near-normal mean values.

Fig. 8

FDG-PET standardized uptake values (SUV) before and after apomorphine treatment. A. Sagittal view of whole-brain SUV averaged over all patients in the apomorphine group (n = 6), demonstrating a predominant increase in posterior areas after treatment compared to before. B. Regional SUV averaged over the 14 brain macrostructures used to design the mixed model, before and after treatment, averaged over all patients in the apomorphine group.