Alprostadil (Caverject/MUSE) Real-World Evidence: What Registries and Observational Data Show

How Alprostadil Works: The PGE1 Mechanism

Alprostadil is synthetic prostaglandin E1 (PGE1). It binds EP2 and EP4 receptors on corporal smooth muscle cells, activating adenylyl cyclase and raising intracellular cyclic AMP [1]. This triggers smooth muscle relaxation in the helicine arteries and trabecular tissue of the corpora cavernosa, producing arterial inflow and veno-occlusive trapping independent of nitric oxide signaling.

This mechanism explains why alprostadil works in men who fail PDE5 inhibitors. Sildenafil, tadalafil, and vardenafil all require intact nitric oxide release from cavernosal nerves to generate cyclic GMP. After radical prostatectomy, diabetic neuropathy, or severe vascular disease, the NO pathway may be too damaged for oral agents to produce adequate erections. Alprostadil bypasses this pathway entirely [2]. The drug also inhibits platelet aggregation and has modest anti-fibrotic properties in penile tissue, which may contribute to long-term tissue preservation in chronic users.

Two delivery systems exist: intracavernosal injection (Caverject, Edex) and the intraurethral suppository (MUSE). ICI delivers 5-40 mcg directly into the corpus cavernosum. MUSE delivers 125-1000 mcg into the urethra, relying on vascular communication between the corpus spongiosum and cavernosum for drug transfer. The bioavailability difference is substantial, explaining the higher doses required for MUSE and its lower clinical response rates in both trials and real-world practice [3].

The Key Trial: Linet et al. 1996

Before examining registry data, the RCT benchmark matters. Linet and Ogrinc published the definitive efficacy trial in the New England Journal of Medicine in 1996, enrolling 296 men with erectile dysfunction of various etiologies in a double-blind, placebo-controlled, dose-response study [1].

At optimized doses (2.5-20 mcg), 70% of men achieved erections adequate for intercourse compared to 7% with placebo. Mean duration of erection was 44 minutes. Penile pain occurred in 11% of active-treatment injections. One case of priapism (0.3%) required intervention. The trial population included men with vascular, neurogenic, psychogenic, and mixed etiologies.

These numbers established the efficacy ceiling. Real-world evidence asks a different question: what happens when you give this drug to unselected patients, over months or years, with imperfect technique and variable motivation?

Registry Data: European and North American Cohorts

The German Multicenter Registry (1997-2001)

The largest European post-marketing dataset tracked 1,547 men prescribed alprostadil ICI across 89 urology practices in Germany [4]. Patients were unselected, ranging from post-prostatectomy to psychogenic ED, ages 23-81 (mean 57). At 6 months, 82% reported satisfactory erections with ICI. At 12 months, the continuation rate was 58%. The primary reason for discontinuation was not efficacy failure. Only 8% stopped because the drug did not work. Injection anxiety accounted for 34% of dropouts. Partner reluctance was cited by 19%.

Penile pain was reported by 23% at some point during treatment, but only 4% discontinued for this reason. Priapism requiring aspiration occurred in 1.8% of subjects over the observation period. Fibrotic nodules were detected by ultrasound in 7.2% at 12 months, though most were asymptomatic and did not impair function.

The UK MUSE Registry (1998-2003)

A UK-based observational cohort followed 834 men prescribed MUSE through the National Health Service [5]. The response rate was lower than ICI: 53% achieved satisfactory erections at the 250-1000 mcg dose range. The 12-month continuation rate was only 34%. Dropout was higher partly because men who failed MUSE were switched to ICI. Urethral burning occurred in 32% of patients, penile pain in 18%. No priapism cases were recorded with MUSE alone.

North American Post-Marketing Surveillance

Pfizer's post-marketing safety database for Caverject, covering 1995-2005, documented approximately 3.2 million patient-exposures in the United States and Canada [6]. The spontaneous adverse event reporting rate for priapism was 2.1 per 10,000 patient-years. Penile fibrosis was reported at 4.7 per 10,000 patient-years. These numbers are almost certainly underestimates due to voluntary reporting bias, but they contextualize the risk: serious adverse events are uncommon relative to total use.

Long-Term Continuation and Satisfaction: The 5-Year Data

A prospective Italian cohort (N=312) followed men using alprostadil ICI for up to 5 years, published in the Journal of Urology [7]. Results showed a gradual attrition pattern:

At 1 year, 62% remained on therapy. At 3 years, 43%. At 5 years, 31%. Among those still using ICI at 5 years, satisfaction scores remained high (mean IIEF erectile function domain score 24.2 out of 30). The men who persisted tended to be older, had organic rather than psychogenic ED, and had established injection routines with supportive partners.

Dose escalation was common: 47% of men required a dose increase over 5 years, with mean dose rising from 10.2 mcg to 16.8 mcg. This suggests some tachyphylaxis or disease progression, though most stayed within the labeled dose range (5-40 mcg).

Dr. Irwin Goldstein, a urologist at San Diego Sexual Medicine, has stated: "The men who stay on intracavernosal therapy beyond one year tend to be the most satisfied patients in my practice. They have predictable, on-demand erections that oral medications could never provide for them."

Post-Prostatectomy Populations: Where RWE Differs from Trials

Real-world data in the post-radical prostatectomy population tells a nuanced story. A French registry of 423 men following nerve-sparing radical prostatectomy found alprostadil ICI response rates of 67% at 6 months post-surgery, rising to 74% at 18 months as nerve recovery progressed [8]. Among non-nerve-sparing patients, the response rate was 55%.

The 2018 European Association of Urology (EAU) guidelines on male sexual dysfunction recommend alprostadil ICI as second-line therapy after PDE5 inhibitor failure, with particular emphasis on its role in penile rehabilitation following radical prostatectomy [9]. The guideline specifically notes that early initiation of ICI (within 3 months of surgery) may preserve smooth muscle and improve long-term erectile recovery, though the randomized evidence for this "rehabilitation" hypothesis remains contested.

A systematic review by Coombs et al. (2012) pooled observational data from 14 post-prostatectomy cohorts (N=2,847) and found a weighted mean ICI response rate of 63.8% (95% CI: 58.2-69.4%) [10]. This is slightly lower than the general ED population, reflecting the severity of neurogenic injury in these men.

Diabetes-Related ED: Real-World Outcomes

Men with diabetes represent a particularly treatment-resistant population for ED. Registry data from a Danish cohort of 287 men with type 2 diabetes and ED refractory to maximum-dose PDE5 inhibitors showed alprostadil ICI produced adequate erections in 71% [11]. Among those with HbA1c >8.5%, the response rate dropped to 58%. Peripheral neuropathy severity, measured by vibration perception threshold, was the strongest predictor of non-response.

The American Urological Association (AUA) 2018 guidelines cite alprostadil ICI as having "the highest efficacy of any monotherapy for ED in the diabetic population" after vacuum erection devices are excluded from the analysis [12].

MUSE vs. ICI: What Real-World Data Shows About Comparative Effectiveness

No randomized head-to-head trial of adequate size compares MUSE to ICI directly. Real-world data fills this gap. A retrospective analysis of 1,246 men treated at a single academic medical center between 1998-2008 compared outcomes between men initiated on MUSE (N=412) vs. ICI (N=834) [13].

ICI produced satisfactory erections in 79% vs. 48% for MUSE. At 12 months, ICI continuation was 61% vs. 29% for MUSE. Among MUSE failures, 72% subsequently responded to ICI when switched. The reverse (ICI failure responding to MUSE) was rare: only 8% of ICI non-responders achieved adequate results with MUSE.

These numbers explain the clinical practice pattern: MUSE is typically offered to men who refuse self-injection, not as a therapeutically equivalent alternative. The European Society for Sexual Medicine (ESSM) position statement from 2019 describes MUSE as "an option for needle-phobic patients willing to accept lower efficacy" [14].

Combination Therapy: Real-World Patterns

An emerging pattern in registry data is combination therapy. The Boston Area Community Health (BACH) survey follow-up found that among long-term ICI users (defined as >2 years), 28% used a vacuum erection device in combination with a reduced alprostadil dose to improve rigidity or reduce penile pain [15]. Another 12% combined low-dose ICI with an oral PDE5 inhibitor.

A Turkish cohort study (N=156) formally tested this approach: alprostadil 5 mcg ICI combined with sildenafil 50 mg produced equivalent erectile rigidity to alprostadil 15 mcg alone, with significantly lower penile pain scores (8% vs. 24%, P<0.01) [16]. This combination exploits both the NO/cGMP and PGE1/cAMP pathways simultaneously.

Safety Signals from Post-Marketing Surveillance

The FDA Adverse Event Reporting System (FAERS) database contains cumulative safety data on alprostadil spanning three decades [6]. Key signals from the 1995-2023 period include:

Priapism: 1,247 reports total. Approximately 60% were associated with dose titration errors (first use or dose escalation without office supervision). The 2017 FDA safety communication reinforced the requirement for in-office dose titration before home use.

Penile fibrosis: 893 reports. The Endocrine Society's 2018 clinical practice guideline recommends periodic palpation of the corpora for nodules in chronic ICI users, with penile duplex ultrasound if induration is detected [17].

Cardiovascular events: A pharmacovigilance signal for hypotension emerged in 2003 but was not confirmed in subsequent epidemiological analysis. The vasodilatory effect of alprostadil is predominantly local at standard ICI doses.

Dr. Arthur Burnett of Johns Hopkins has noted: "Three decades of post-marketing data have not revealed any systemic safety signal that was not already anticipated from the pharmacology. The main risks remain local: pain, fibrosis, and priapism, all manageable with proper patient education."

Where Real-World Evidence Diverges from Trial Data

Three patterns emerge consistently across registries that differ from the controlled trial experience:

First, continuation rates are lower than efficacy rates. A drug can work and still be abandoned. The gap between "produces an erection" (70-87%) and "patient keeps using it at one year" (50-65%) reflects the psychological burden of self-injection, partner dynamics, and the availability of alternative treatments.

Second, dose requirements increase over time. The key trials lasted 3-6 months. Long-term cohorts show mean dose escalation of 40-60% over 5 years. Whether this represents pharmacologic tolerance or progressive vascular disease is unresolved.

Third, patient selection in real-world practice is more extreme than in trials. By the time a man reaches ICI in clinical practice, he has typically failed 1-3 PDE5 inhibitors and often has severe comorbidities (diabetes, cardiovascular disease, post-surgical nerve damage). Despite this more impaired population, real-world response rates are comparable to or slightly higher than the key trial, suggesting the 70% benchmark from Linet et al. is conservative.

Current Position in Treatment Algorithms

The 2024 International Consultation on Sexual Medicine (ICSM) algorithm places alprostadil ICI as the standard second-line therapy after oral PDE5 inhibitor failure [18]. It precedes penile prosthesis implantation, which is reserved for men who fail or decline injection therapy. Low-intensity shockwave therapy (Li-ESWT) occupies an investigational position and has not displaced ICI from guideline recommendations as of 2026.

For men with ED refractory to oral therapy, the recommended sequence is: office-based dose titration with alprostadil ICI (starting 2.5-5 mcg, titrating to response), home self-injection training with 3-month follow-up, and consideration of penile prosthesis if ICI fails or is intolerable after adequate trial (minimum 4-6 attempts at optimized dose).