Analysis of FDA reports finds Prozac’s sexual side effects vary widely, with some emerging years into treatment

A recent study published in Progress in Neuropsychopharmacology & Biological Psychiatry suggests that sexual side effects associated with the antidepressant fluoxetine are highly varied, involving severe sensory disruptions alongside changes in libido. Using computer modeling, the researchers found that the drug may interfere with immune and cellular stress proteins, offering clues about the biological mechanisms behind these symptoms.

Fluoxetine is a widely prescribed medication used to treat depression and anxiety. It belongs to a class of drugs called selective serotonin reuptake inhibitors, or SSRIs. These medications work by increasing the availability of serotonin, a chemical messenger in the brain that helps regulate mood. While effective for mental health conditions, these drugs frequently produce unwanted sexual side effects.

Lead authors Jun Li and Xing Wang, along with corresponding author Honglin Hu from Nanchang University in China, initiated the investigation to better understand these adverse reactions. Previous research pointed to serotonin as the primary culprit, as heightened serotonin activity can suppress other brain chemicals like dopamine, which drives sexual arousal. The research team suspected that other biological networks, such as inflammation and cell survival pathways, might also be involved in these side effects. They combined real-world patient reports with advanced computer simulations to explore this hypothesis.

The scientists began by extracting data from the FDA Adverse Event Reporting System. This public database collects voluntary reports of adverse drug reactions from patients and healthcare professionals. The team filtered the records from 2004 to 2025 to isolate cases where fluoxetine was listed as the primary suspected drug. They analyzed 997 reports related to sexual dysfunction to see if certain side effects were disproportionately associated with the medication.

The analysis revealed strong statistical signals for a wide range of sexual problems. Common issues like erectile dysfunction and decreased libido appeared frequently in the patient reports. However, the researchers observed the strongest reporting intensities for highly specific conditions. Symptoms such as a total inability to climax, decreased orgasmic sensation, and genital numbness showed unusually high reporting rates compared to traditional measures of sexual dysfunction.

Free daily newsletter

Following the database analysis, the team calculated the time it took for these symptoms to appear after a patient started taking the medication. They only included reports that contained complete date information for both the start of treatment and the onset of the side effect. They used statistical distribution models to map out the timing patterns for different symptoms across the patient population.

The timing of the side effects varied greatly depending on the specific symptom. Some conditions, such as prolonged and painful erections or persistent genital arousal, emerged within the first few days of treatment. Issues related to sexual desire, such as a loss of libido, tended to surface after about four to five months. The most delayed side effects involved sensory and orgasmic disruptions, with reports of genital numbness and an inability to climax often appearing after more than two years of continuous medication use.

To understand the biological roots of these symptoms, the researchers turned to network toxicology. This computational approach maps out how drugs interact with the body by cross-referencing vast genetic and protein databases. The team first compiled a list of human proteins known to interact with fluoxetine. They then created a separate list of proteins and genes associated with human sexual dysfunction.

By comparing the two lists, the scientists identified 56 overlapping genetic targets. These shared targets operate within networks that control brain chemistry, including both serotonin and dopamine pathways. The overlapping targets were also heavily concentrated in biological pathways that regulate inflammation, immune responses, and cellular death.

The researchers built a digital map to visualize how these 56 targets interact with one another inside the body. This mapping process identified two core hub proteins that act like major intersections in a city traffic grid, handling a high volume of biological signals. The first is IL10, a protein that helps regulate the body’s inflammatory and immune responses. The second is TP53, a protein that responds to cellular stress and helps control when a damaged cell undergoes programmed death.

To see if the drug could physically influence these proteins, the team used a technique called molecular docking. This computer simulation predicts how a small drug molecule might fit into the three-dimensional structure of a larger protein. It operates much like a simulation of a key entering a lock.

The simulations showed that the fluoxetine molecule fits neatly into the structural pockets of both the TP53 and IL10 proteins. The binding affinity was notably strong for the IL10 protein. This physical compatibility suggests that the drug might directly attach to and alter the function of these proteins in the human body. The researchers suspect that fluoxetine might contribute to sexual side effects by triggering low-level inflammation or cellular stress in nerve and reproductive tissues.

Because the FDA database relies on voluntary submissions, it cannot capture the true frequency of these side effects in the general population. The reporting system also lacks complete information on drug dosages, the duration of treatment, and the patients’ other medications. These missing details prevent researchers from determining exactly how much of the drug is required to trigger a specific symptom.

The database does not track the severity of a patient’s underlying depression. Since depression itself is associated with bodily inflammation, it is difficult to isolate the medication as the sole driver of the immune system changes observed in the data. The underlying mental health condition may be contributing to the same biological pathways highlighted in the computer models.

The computer models used to predict protein interactions are entirely theoretical. While molecular docking shows that a physical bond is possible, it does not confirm how the drug behaves inside a living human body. Physical laboratory experiments and clinical trials are necessary to verify these biological mechanisms.

The study, “Deciphering the molecular mechanisms of fluoxetine-induced sexual dysfunction: Evidence from pharmacovigilance, network toxicology, and molecular docking,” was authored by Jun Li, Xing Wang, Yunfeng Zhang, Penglin Zhang, Zhenfa Tan, Ruizhen Huang, Zuhuan Xu, Zhiyu Zhou, and Honglin Hu.