Multimodal prognosis of negative symptom severity in individuals at increased risk of developing psychosis

Abstract

Negative symptoms occur frequently in individuals at clinical high risk (CHR) for psychosis and contribute to functional impairments. The aim of this study was to predict negative symptom severity in CHR after 9 months. Predictive models either included baseline negative symptoms measured with the Structured Interview for Psychosis-Risk Syndromes (SIPS-N), whole-brain gyrification, or both to forecast negative symptoms of at least moderate severity in 94 CHR. We also conducted sequential risk stratification to stratify CHR into different risk groups based on the SIPS-N and gyrification model. Additionally, we assessed the models' ability to predict functional outcomes in CHR and their transdiagnostic generalizability to predict negative symptoms in 96 patients with recent-onset psychosis (ROP) and 97 patients with recent-onset depression (ROD). Baseline SIPS-N and gyrification predicted moderate/severe negative symptoms with significant balanced accuracies of 68 and 62%, while the combined model achieved 73% accuracy. Sequential risk stratification stratified CHR into a high (83%), medium (40-64%), and low (19%) risk group regarding their risk of having moderate/severe negative symptoms at 9 months follow-up. The baseline SIPS-N model was also able to predict social (61%), but not role functioning (59%) at above-chance accuracies, whereas the gyrification model achieved significant accuracies in predicting both social (76%) and role (74%) functioning in CHR. Finally, only the baseline SIPS-N model showed transdiagnostic generalization to ROP (63%). This study delivers a multimodal prognostic model to identify those CHR with a clinically relevant negative symptom severity and functional impairments, potentially requiring further therapeutic consideration.

Fig. 1. Model performances.

A Posterior balanced accuracy distributions of baseline negative symptom (upper panel), gyrification (middle panel), and combined model (lower panel). Shaded grey area indicates 95% of the probability mass of the respective posterior distribution over the balanced accuracy. B Confusion matrices and C receiver operating characteristic (ROC) curves of the prediction models. Bad outcome: Expression of moderate to severe negative symptoms at follow-up (any score ≥3). Good outcome: All SIPS negative items <3 at follow-up. SIPS-N negative symptoms measured the Structured Interview for Psychosis-Risk Syndromes .

Fig. 2. Feature importance.

A Feature importance of baseline negative symptom model. Feature importance was measured through cross-validation (CV) ratio profiles (CVR = mean(w)/SE(w), where w corresponds to the normalized weight vector under Euclidian assumptions of the logistic regression or support vector machine classifier; see ref. for more details). Negative CVRs indicate that reduced values of the predictor are associated with increased risk of expressing negative symptoms, whereas positive values imply that an increase of the predictor value is associated with increased risk. B Top 10 most important features of the gyrification model. C Probabilistic assessment diagram illustrating two-stage sequential risk stratification to stratify patients based on their risk to develop moderate to severe negative symptoms. Note that this computation is based on (1) the base rate, as well as sensitivity and specificity from (2) the baseline SIPS-N and (3) the gyrification model, all derived from our sample. X-axis: Sequential tests (based on baseline negative symptoms and gyrification). Y-axis: Positive predictive value (PPV) associated with expressing moderate/severe negative symptoms at 9 months follow-up. The pretest probability was set to 40% based on our own sample (see Table 1). Each bifurcation in the plot represents the updated PPV after knowing that a test either yielded a positive (ascending solid line) or a negative result (descending dashed line). For a schematic analysis overview, please refer to Fig. S1. Line colour: Level of risk as previously suggested. High: >80%, medium: 40–64%, and low: <20%. Dot sizes: Relative proportion of participants in our sample with a corresponding number of positive tests. The diagram also illustrates the number needed to treat (NNT) at each node, which is based on the risk ratio of a recent clinical pilot trial with a d-serine intervention to treat negative symptoms in clinical high-risk individuals. SIPS-N negative symptoms measured the Structured Interview for Psychosis-Risk Syndromes.