Tapering the antipsychotics: finding the correct approach

Antipsychotics are recommended for long -term treatment of schizophrenia because they reduce the risk of relapse. However, antipsychotics have many adverse effects, including metabolic complications, tardive dyskinesia, and probable brain volume reduction. There is some evidence that not all patients need life-long antipsychotic treatment and some may have improved social functioning when taking less or no antipsychotic. There are currently no published guidelines on reduction or cessation of an antipsychotic; thus the following principles proposed are relevant while deprescribing is thought appropriate.

Relapse Related to Withdrawal

Relapse often occurs when antipsychotics are withdrawn. The nature of the process of withdrawing antipsychotics may itself be associated with relapse, other than representing an unmasking of underlying chronic illness. This is evidenced by the occurrence of psychotic symptoms in people without a psychotic disorder who abruptly stop taking antipsychotics used to treat other conditions, such as nausea or lactation difficulties. Likewise, it is one possible reason for the marked preponderance of relapses soon after abrupt antipsychotic cessation in patients with schizophrenia in discontinuation trials. In one meta-analysis, it was found that 60% of all relapses over 4 years occurred within 3 months of drug cessation.

Underlying Neurobiology- dopaminergic hypersensitivity

Withdrawal-associated relapse has been attributed to neural adaptations to long-term antipsychotic treatment (dopaminergic hypersensitivity) that persist after antipsychotic cessation. Indeed, molecular imaging studies in individuals with schizophrenia have found increased D2 /D3 receptor availability in those who had been exposed to antipsychotic medication but not in antipsychotic-naive patients. This hypersensitivity to dopamine may render patients more susceptible to psychotic relapse when D2 blockade is diminished by antipsychotic dose reduction. Neuroadaptive effects of antipsychotic treatment can persist for months or years after stopping.

Tardive dyskinesia—attributed to dopaminergic hypersensitivity—can persist for years after antipsychotic medication has been ceased. There is also evidence that patients who have discontinued antipsychotics have increased rates of relapse for 3 years compared with people who maintain antipsychotic treatment, after which relapse rates converge, suggesting that adaptations may have resolved by this point.

The risk of relapse on cessation of antipsychotics might be minimized by more gradual dose tapering because these neuroadaptations would then have time to resolve during the tapering process and the rate of decline of blockade is more modest. A small meta-analysis found that tapering over 3 to 9 months halved the rate of relapse compared with abrupt discontinuation. Tapering over 4 weeks showed no difference from abrupt discontinuation.

How to taper?

A hyperbolic association between dose of antipsychotic and its therapeutic effects (as measured by symptoms scales) shows, that clinical response mirrors the neurobiological pattern of effects. For example, a reduction from 4 mg to 3 mg to 2 mg to 1 mg to 0 mg of risperidone, is the regimen for linear reduction of anti-psychotic dose. (Figure) Although this regimen appears reasonable, the hyperbolic association between dose and effect on D2 blockade dictates that these linear dose decreases will produce increasingly larger reductions of D2 blockade (and the clinical consequences of this). Indeed, the reduction of dose from 1 mg to 0 mg will produce a reduction in D 2 blockade (44.6%) larger than that produced by the reduction from 10 mg to 1 mg of risperidone (38.8%). These increasingly large reductions in D 2 blockade may be more likely to provoke relapse. These hyperbolic reductions are approximated by sequential halving of dose: for example, risperidone doses of 8mg, 4 mg, 2 mg, 1 mg, 0.5 mg, 0.25 mg, 0.125 mg, and 0 mg produce roughly 10% point reductions in the extent of D 2 blockade. This pattern of reduction may be less likely to provoke relapse because it avoids large increases in dopaminergic signalling.

To summarise, when antipsychotic medication reduction is appropriate, it should be done very slowly (over months or years) and in a hyperbolic manner. For example, the dose could be reduced by an amount calculated to reduce 10% points of the drug's D2 blockade (approximately equal to a reduction in half the last dose given) every 3 to 6 months. Final doses before complete cessation will need to be very small to prevent a large decrease in D 2 blockade. This may need to be as small as 2.5% the original therapeutic dose. Delivery of these small doses may require splitting tablets or using liquid formulations. Such a reduction regimen might reduce the risk of relapse on discontinuation.

Source: JAMA Psychiatry: Horowitz MA, Murray RM, Taylor D. Tapering Antipsychotic Treatment. JAMAPsychiatry. 2021;78(2):125–126. doi:10.1001/jamapsychiatry.2020.2166