Low-Risk and Sweet-Spot Second-Generation Antipsychotics Identified Through Receptor Pharmacology and Cardiometabolic TWAS

<title>Abstract</title> <p> <bold>Background</bold> Metabolic adverse effects remain a major limitation of second-generation antipsychotic prescribing. Weight gain, dyslipidemia, hypertension, and type 2 diabetes can affect long-term adherence and may also contribute indirectly to kidney disease risk. Zhou et al. (2024) recently showed that kidney stones and chronic kidney disease share strong cardiometabolic comorbidities, particularly hypertension, type 2 diabetes, and obesity, although common kidney stones were unlikely to be an independent direct cause of CKD. This makes cardiometabolic burden a clinically relevant axis when choosing antipsychotics for patients with renal or metabolic vulnerability. <bold>Objective</bold> This study ranked N05A antipsychotics, with emphasis on second-generation agents, by predicted metabolic liability using receptor affinity, defined daily dose, literature-derived metabolic receptor weights, and TWAS signals from Zhou-linked CKD and cardiometabolic traits. The primary aim was to identify low-risk and “sweet-spot” antipsychotics that combine low predicted metabolic burden with meaningful DRD2 engagement. <bold>Methods</bold> A Ki–DDD–TWAS pipeline was applied to N05A drugs. Receptor affinity was dose-weighted using the formula affinity_dose_score = (1 / Ki_nM) × log(DDD_mg + 1). Scores were then weighted by metabolic relevance, with highest weights assigned to HRH1, HTR2C, and CHRM3. TWAS data were integrated across four Zhou-linked traits: CKD, hypertension, type 2 diabetes, and obesity. Downstream analyses assessed cross-trait stability, receptor-burden decomposition, low-risk status, DRD2 sweet-spot status, residual TWAS signals, and concordance with clinical antipsychotic metabolic-risk literature. <bold>Results</bold> Risk rankings were highly stable across traits, with mean off-diagonal Spearman rho of 0.996. Forty-two drugs met robust low-risk criteria. Among clinically important second-generation antipsychotics, aripiprazole, lurasidone, brexpiprazole, cariprazine, quetiapine, iloperidone, amisulpride, and paliperidone fell within the low-risk range, although some retained gene-specific TWAS signals. Thirty-five drugs met sweet-spot criteria, including aripiprazole, brexpiprazole, cariprazine, lurasidone, quetiapine, iloperidone, and amisulpride. Low-risk drugs showed markedly lower HTR2C, CHRM3, and HRH1 burden than high-risk drugs. High-risk agents included clozapine, asenapine, zotepine, ziprasidone, risperidone, olanzapine, and sertindole, with chlorpromazine, thioridazine, and chlorprothixene providing first-generation context. <bold>Conclusions</bold> This exploratory receptor pharmacology–TWAS framework identified several low-risk and sweet-spot second-generation antipsychotics that may be clinically useful when metabolic or CKD risk is a priority. The model aligned well with clinical literature for clozapine and olanzapine but diverged for some agents, especially ziprasidone, risperidone, and asenapine. These differences were mechanistically interpretable and support using the framework as a decision-support layer, not as a replacement for patient-level evidence. </p>