Factors predicting symptom stability without antipsychotic treatment in schizophrenia: analysis of 557 patients on placebo for at least 6 months

Background

The optimal duration of antipsychotic treatment in individuals with chronic schizophrenia remains a subject of ongoing debate. Continuous antipsychotic maintenance is widely regarded as the standard of care for relapse prevention; however, this approach is associated with long-term adverse effects and challenges with medication adherence. An alternative strategy—intermittent antipsychotic treatment, whereby medication is administered during periods of symptom exacerbation and discontinued once clinical stability is achieved—has been proposed as a means of reducing cumulative drug exposure while potentially improving tolerability and adherence. Evidence suggests that not all patients derive equal benefit from continuous treatment, underscoring the need to identify clinical or demographic factors that might predict which individuals can remain stable without ongoing medication. Such predictors could inform more personalized and flexible treatment strategies.

Objectives

The primary objective was to identify baseline characteristics that predict the ability to maintain symptomatic stability in the absence of continuous antipsychotic therapy. Building on prior findings [1,2], we hypothesized that individuals with predominantly negative symptom profiles and low levels of agitation at baseline would demonstrate a lower risk of relapse during placebo treatment and may therefore be suitable candidates for intermittent maintenance strategies.

Methods

We conducted a participant-level meta-analysis using data from five randomized, double-blind, placebo-controlled antipsychotic trials that included maintenance phases of at least six months. Data were obtained through the YODA and Vivli data-sharing platforms. Primary outcomes were time to relapse—defined as a ≥12-point increase in the Positive and Negative Syndrome Scale (PANSS) [1] —as well as treatment discontinuation and psychiatric hospitalization. Secondary outcomes examined relapse timing based on Clinical Global Impression–Severity (CGI-S) and Personal and Social Performance (PSP) scores. Predictor variables included baseline PANSS item scores, CGI-S, PSP, demographic characteristics, smoking status, age at illness onset, hospitalization history, prior antipsychotic exposure, and the presence of tardive dyskinesia or akathisia. Time-to-event outcomes were analyzed using multivariable Cox proportional hazards models, with random-effects meta-analytic techniques applied to account for between-study heterogeneity. Study name was included as a control variable, and trial inclusion criteria and baseline patient characteristics were systematically compared prior to data pooling.

Results

A total of 557 participants across five studies who were randomized to placebo for at least six months were included. Higher baseline positive symptom severity was consistently associated with an increased risk of relapse, whereas greater baseline negative symptom severity was associated with longer relapse-free survival. Analyses using dichotomized negative symptom groups, consistent with prior classifications, and interaction models supported this pattern. Network analyses identified symptom clusters associated with more stable outcome trajectories, and propensity score methods were used to refine predictors of suitability for intermittent maintenance. Additional clinical variables, including smoking status, extrapyramidal symptoms, and medical history were examined.

Conclusions

These findings suggest that prominent negative symptoms at baseline may predict greater clinical stability during periods without antipsychotic treatment. This supports the potential feasibility of intermittent maintenance strategies in carefully selected patients and advances efforts toward more individualized, risk-adapted treatment approaches that balance relapse prevention with long-term tolerability and quality of life.

References

1. Rabinowitz, J., Levine, S.Z., Garibaldi, G., Bugarski-Kirola, D., Berardo, C.G., Kapur, S., 2012. Negative symptoms have greater impact on functioning than positive symptoms in schizophrenia: analysis of CATIE data. Schizophrenia research 137(1-3), 147–150.
2. Rabinowitz, J., Staner, C., Saoud, J., Weiser, M., Kuchibhatla, R., Davidson, M., Harvey, P.D., Luthringer, R., 2023. Long-term effects of Roluperidone on negative symptoms of schizophrenia. Schizophrenia Research 255, 9–13.
3. Siafis, S., Brandt, L., McCutcheon, R.A., Gutwinski, S., Schneider-Thoma, J., Bighelli, I., Kane, J.M., Arango, C., Kahn, R.S., Fleischhacker, W.W., McGorry, P., Carpenter, W.T., Falkai, P., Hasan, A., Marder, S.R., Schooler, N., Engel, R.R., Honer, W.G., Buchanan, R.W., Davidson, M., Weiser, M., Priller, J., Davis, J.M., Howes, O.D., Correll, C.U., Leucht, S., 2024. Relapse in clinically stable adult patients with schizophrenia or schizoaffective disorder: evidence-based criteria derived by equipercentile linking and diagnostic test accuracy meta-analysis. The Lancet Psychiatry 11(1), 36–46.