Summary

• Pituitary adenoma has been reported since market introduction.¹
• In a pharmacovigilance study by McCarren et al (2012), risperidone did not appear to have an increased risk in the incidence of medical records validated pituitary tumors with mass effects compared to other atypical agents (event rates: 2.5/100,000 person-years, each).²
• Szarfman et al (2006)³ analyzed the disproportionality of reporting of pituitary tumors, hyperprolactinemia, and galactorrhea in patients treated with risperidone, olanzapine, quetiapine, ziprasidone, aripiprazole, and haloperidol using the FDA’s Adverse Events Reporting System (AERS) database between January 1968 and May 2005. The highest adjusted reporting ratios for pituitary tumor, hyperprolactinemia, and galactorrhea were with risperidone. These data may represent an adverse event signal, but do not establish a causal relationship.
• Several case reports^4–16 have identified pituitary adenomas detected in patients while on risperidone, with authors recommending that further research is needed to explore whether a relationship between pituitary tumors and antipsychotics may exist in humans. In most cases, clinical symptoms potentially related to hyperprolactinemia resulted in an endocrinologic evaluation, including MRI (magnetic resonance imaging) or CT (computerized tomography), for assessment of pituitary abnormalities.

BACKGROUND

Pituitary tumors commonly occur in the general population (10-25% incidence) and are located beneath the base of the brain; 99% are benign (McDowell¹⁷; The Neuroendocrine Clinical Center and Pituitary Tumor Center¹⁸).  Pituitary tumors are sometimes serendipitously noted during a CT or MRI exam that was performed for other reasons (Mavrakis 2004)¹⁹.  Although the majority of pituitary tumors are small (less than 10 mm in greatest diameter) and clinically silent, some pituitary tumors may be hormonally active and cause associated symptoms (e.g. prolactin production) or grow large and cause symptoms by compressing neighboring structures (“mass effect”). Symptoms include headaches (which can occur with small or large tumors) and vision problems.

PRODUCT LABELING

Please refer to the following sections of the enclosed Full Prescribing Information that are relevant to your inquiry: WARNINGS AND PRECAUTIONS and ADVERSE REACTIONS.

CLINICAL DATA

Double-Blind Trials

Koenigsberg et al (2003)²⁰ conducted a 9-week, double-blind, randomized, placebo-controlled study to evaluate the efficacy and safety of low dose risperidone (starting dose 0.25 mg/day, titrate up to 2 mg/day) in 25 patients with schizotypal personality disorder

(SPD). Patients were evaluated with the Positive and Negative Syndrome Scale (PANSS), Schizotypal Personality Disorder Questionnaire, the Hamilton Rating Scale for Depression, and the Clinical Global Impressions scale.

• Overall, low-dose risperidone was effective in reducing symptom severity and was well tolerated in these patients.
• One patient developed galactorrhea at a 1-mg dose, and her risperidone dose was lowered to 0.5 mg/day and then 0.25 mg/day before she was terminated from the study. Endocrinologic evaluation later revealed a pituitary microadenoma.

Pharmacoepidemiologic Studies

Janssen Research and Development designed the protocol for a pharmacoepidemiologic study (with agreement from FDA on the definition of the primary outcomes) to examine the relationship between risperidone and pituitary tumors with mass effect. McCarren et al (2012)² utilized Optum Insight Life Sciences Research Database (LSRD; 01 July 2002 to 31 December 2007) and the Department of Veterans Affairs (VA) medical care databases (01 June 1999 to 30 September 2009), to compare the association of risperidone (and other atypical antipsychotics) with pituitary tumors with mass effects.

Study Design/Methods

• Retrospective, nested, case-control study with a cohort of patients treated with ≥2 consecutive (≤3 months between prescriptions) antipsychotic dispensations (risperidone versus other atypicals versus conventional antipsychotics).
  • Exclusion criteria included: pharmacy claim for prescribed drug 6 months prior to first dispensation; medical claim consistent with malignancy or pituitary tumor prior to first drug exposure
  • Cohort follow-up period began at the index date (start of second dispensation) and lasted until censoring event (pituitary tumor diagnosis; last claim in database; 9 months before a pregnancy code) or end of study, whichever came first.
• Pituitary tumor with mass effects: Medical diagnosis of a pituitary tumor plus one of the following: pituitary surgery, pituitary radiation therapy, visual field disturbances or pituitary hormone deficiency.
• Primary cohort analysis: Incidence rates of pituitary tumors with mass effects (confirmed by medical records) between risperidone and other atypicals (cox proportional-hazards regression model; adjusted for age only due to limited number of confirmed cases).
  • Secondary outcomes:
    • Case-control analysis: Calculated ORs for pituitary tumors with mass effects (confirmed by medical records) comparing risperidone to other atypicals (conditional logistic regression model).

Results

• In patients treated with atypical antipsychotics, the incidence of validated pituitary tumors with mass effects was comparable to that of the general population (1.4 -2.5 per 100,000 person-years versus 2.8 per 100,000 person-years, respectively) (Table: Pituitary Tumor Analyses – VA and LSRD Databases).
  • Compared to other atypical agents, risperidone did not appear to have an increased risk in the incidence of medical records validated pituitary tumors with mass effects.

EudraVigilance European Pharmacovigilance Database

Lertxundi et al (2018)²¹ analyzed the relationship between atypical antipsychotics and pituitary tumors in the EudraVigilance European pharmacovigilance database.

Methods/Results

• The search included all suspected spontaneous cases of pituitary tumors associated with atypical antipsychotics up to 23 March 2017 in EudraVigilance. The association between tumor and drug was assessed by calculating proportional reporting ratios (PRRs).
• After deduplication, 292 cases were assessed (72% were women; median age was 36 [range, 5-73] years).
  • PRR was the highest for amisulpride (51.57; 95% CI, 36.3-73.2), followed by risperidone (21.83; 95% CI, 18.4-25.8) and its metabolite paliperidone (19.95; 95% CI, 14.7-27.1).
  • Risperidone was the most frequently implicated drug in generating a safety signal, with 156 reported cases, of which 24 were reported with the depot formulation.

Claims-based Study

Gianfrancesco et al (2009)²² discussed the potential bias in testing for hyperprolactinemia and pituitary tumors in risperidone-treated patients in a claims-based study.

Methods/Results

• Two databases (large commercial; Medicaid) were utilized to identify treatment episodes with risperidone, clozapine, olanzapine, quetiapine, ziprasidone, aripiprazole, haloperidol, perphenazine, and “other typical” (reference group) antipsychotics.
• Analysis of 197,926 treatment episodes revealed that among the antipsychotics, risperidone was associated with a significantly greater likelihood of prolactin assessment in patients with or without preceding prior potentially prolactin-related adverse events (Hazard Ratio (HR): 1.34, 95% Confidence Interval (CI): 1.09-1.66; p=0.007).
• A significantly higher likelihood of an MRI/CT was observed in patients with hyperprolactinemia or prior potentially prolactin-related adverse events receiving risperidone (OR (odds ratio): 1.66, 95% CI: 1.23-2.23; p=0.001) or ziprasidone (OR: 1.66, 95% CI: 1.06-2.62; p=0.028).
• The authors concluded that the disproportionate identification and reporting of pituitary adenomas in risperidone-treated patients may be a result of more evaluations in this patient population.

WHO ADR Database

Doraiswamy et al (2007)²³ reviewed reports of pituitary neoplasms in the WHO ADR (adverse drug reaction) database for risperidone and amisulpride.

Methods/Results

• The authors generated two IC (information component) scores:
  • IC (a measure of the strength of the dependency between a drug and an ADR)
  • IC025 (value of the lower 95% confidence limit for the IC)
• In the WHO ADR database, risperidone had the second highest IC score for pituitary neoplasm benign (IC=4.03, IC025=3.33, 19 reports) and pituitary neoplasm NOS (IC=4.49, IC025=3.86, 23 reports).
• The authors noted that IC scores do not indicate causality, but provide a measure of a disproportional relationship of a drug-ADR combination within the database.

FDA Adverse Events Reporting System (AERS) Database

Szarfman et al (2006)³ analyzed the disproportionality of reporting of pituitary tumors, hyperprolactinemia, and galactorrhea in patients treated with risperidone, olanzapine, quetiapine, ziprasidone, aripiprazole, and haloperidol using the FDA’s AERS database between January 1968 and May 2005.

Methods

• The Multi-item Gamma Poisson Shrinker (MGPS) statistical algorithm was applied to detect higher than expected drug-event associations (“signals”) within the AERS database.
• The authors searched the database for the following event codes: pituitary tumor, prolactinoma, pituitary tumor benign, blood prolactin increased, hyperprolactinemia, gynecomastia, breast hypertrophy, amenorrhea, and galactorrhea.
• Estimated prescription data from launch of antipsychotic drug through June 2005 was obtained from the IMS National Prescription Audit Plus database.  No controls were used for the length of time of commercial availability for each of the medications.

Results

• Seventy-seven pituitary tumor reports were identified, with 54 reports associated with risperidone.  Women accounted for 82% of all reports.
• Risperidone had the highest adjusted reporting ratios (disproportionality signal scores) for pituitary tumors, hyperprolactinemia, and galactorrhea, followed by haloperidol, ziprasidone, and olanzapine.
• Table: Number of Reported Pituitary Tumors displays the number of pituitary tumors reported with each of the seven antipsychotics evaluated.
  • The numbers reported below account for patients who may have been using more than one antipsychotic at a time, though only 20% of risperidone reports noted concomitant use of another antipsychotic or other suspect drug.
  • Approximately 50% of the tumors were identified as benign, with pituitary tumor complications reported with risperidone (n=14), haloperidol (n=4), clozapine (n=3), and olanzapine (n=1).

• Three adolescents (age 14-16 years) treated with risperidone, and 1 child/adolescent treated with olanzapine were reported to have a pituitary tumor.  While reports of hyperprolactinemia or galactorrhea occurred more frequently with risperidone as compared to the other antipsychotics in younger children (age <7 years), more than 90% of cases were in adolescents.
• Data from the IMS National Prescription Audit Plus showed that ziprasidone (6/5,672,000) had the highest total number of reported pituitary tumors relative to prescription volume, followed by risperidone (43/80,987,000), olanzapine (7/62,795,000), and others.

Limitations

• The authors identified several key study limitations:
  • Risperidone’s known association with elevation of prolactin levels may have led to more frequent pituitary investigations, resulting in a detection bias.
  • The AERS database search cannot establish causality; thus, the results of this analysis cannot be used for this purpose and are preliminary.
  • Further investigation, using additional methods, is required to better determine why the numerical differences between the atypical antipsychotic agents were observed.
  • Pharmacovigilance reporting may be influenced by detection and reporting biases inherent in the AERS database.
• Additional AERS database limitations included:
  • Changes in reporting and collection methodology over time
  • Differential underreporting
  • Potential differences in co-morbidities or concomitant medications among exposed patients
  • Inability to completely eliminate duplicate reporting
  • Incomplete case histories.
• The authors concluded that treatment with antipsychotics may be associated with pituitary tumors, but additional epidemiologic studies, long-term comparative studies, and detailed analyses of pituitary tumor cases associated with antipsychotics are necessary to ascertain causality and relative risk.

Case Reports

Additional case reports prior to 2007 have been identified during a literature search and are referenced.^9–15

Review Articles

Review articles, identified during a literature search, have been referenced for your convenience.^24–26

Literature Search

A literature search of MEDLINE®, EMBASE®, BIOSIS Previews®, and DERWENT Drug File (and/or other resources, including internal/external databases) pertaining to this topic was conducted on 15 September 2023.