Viagra vs Alprostadil (Caverject/MUSE): Long-Term Durability of Response

How Each Drug Works and Why Mechanism Predicts Durability

Sildenafil blocks phosphodiesterase type 5, preserving cyclic GMP that would otherwise be degraded, which prolongs smooth-muscle relaxation in the corpora cavernosa. That pathway depends on nitric oxide release from endothelial and neuronal sources. Alprostadil, a synthetic prostaglandin E1 analogue, binds EP receptors directly and raises intracellular cyclic AMP, bypassing the nitric-oxide step entirely. That single mechanistic difference is why alprostadil often works when sildenafil does not, and why the two drugs may be combined in refractory cases. [1]

Sildenafil's Mechanism and Its Limits

The cyclic-GMP pathway sildenafil depends on becomes less functional as endothelial disease progresses. Men with severe diabetes, radical prostatectomy with nerve sacrifice, or advanced vascular disease tend to see sildenafil response erode over time, not because the drug loses potency, but because the upstream signal it amplifies grows weaker. A 2002 review in the Journal of Urology confirmed that post-prostatectomy patients without nerve-sparing had significantly lower sildenafil response rates than nerve-spared peers. [2]

Alprostadil's Mechanism and Its Limits

Alprostadil's cAMP pathway is relatively preserved even in severe vascular or neurogenic ED. The practical ceiling is delivery, not pharmacology. Intracavernosal injection places drug exactly where it is needed; MUSE relies on urethral absorption, which is variable, and effective transfer to the corpus cavernosum depends on intact venous communication between the urethra and the erectile bodies. [3] That absorption variability explains much of MUSE's lower efficacy compared with Caverject.

Key Trial Efficacy: What the Foundational NEJM Data Show

Both drugs cleared regulatory approval based on large, randomized, placebo-controlled trials published in the New England Journal of Medicine, allowing a reasonably direct baseline comparison.

Goldstein et al. 1998: Sildenafil's Registration Trial

Goldstein et al. Enrolled 861 men with erectile dysfunction of mixed etiology in a 24-week double-blind trial. [4] Sildenafil 25 to 100 mg produced improved erections in 69% of participants vs 22% on placebo (P<0.001). Mean scores on the International Index of Erectile Function improved by 7.5 points from baseline in the sildenafil group. Organic etiologies, including diabetes and spinal cord injury subgroups, showed lower but still significant benefit.

Linet et al. 1996: Caverject's Registration Trial

Linet et al. Enrolled 296 men in a double-blind, placebo-controlled crossover study of intracavernosal alprostadil 2.5 to 20 mcg. [5] Successful intercourse occurred after 94% of active injections vs 11% of placebo injections. The dose-response was steep: men using 20 mcg had the highest success rates, while the 2.5 mcg starting dose was frequently insufficient. This steep dose-response has important implications for real-world use, since under-dosing is a common cause of early discontinuation.

What These Trials Do Not Tell Us

Neither the Goldstein nor the Linet trial extended beyond 6 months, so neither directly answers the question of long-term durability. The subsequent 5-year open-label extensions and real-world cohort studies fill that gap, and they tell a different story than the key efficacy numbers suggest.

Long-Term Durability of Sildenafil: 5-Year Evidence

Sildenafil's durability record is stronger than alprostadil's on most metrics, driven primarily by its oral route of administration. A 5-year open-label extension study following men who responded to sildenafil in randomized trials found that 64% of men who had responded at 12 weeks continued to report satisfactory sexual activity at 60 months, with no evidence of pharmacologic tachyphylaxis to the drug itself. [6]

Dropout vs Efficacy Loss: An Important Distinction

Long-term dropout from sildenafil is often attributed to loss of response, but the etiology is more complex. A population-based study in BJU International found that among men who discontinued PDE5 inhibitors, only 28% cited inadequate efficacy as the primary reason. Relationship factors, cost, and loss of partner accounted for the remainder. [7] This means the drug's pharmacologic durability is likely higher than discontinuation rates imply.

Diabetes and Post-Prostatectomy Subgroups

Men with type 2 diabetes experience measurably lower long-term sildenafil response rates. A study of 268 diabetic men using sildenafil for up to 2 years found that 57% maintained satisfactory responses, compared with roughly 80% in non-diabetic cohorts. [8] Post-prostatectomy men with nerve-sparing procedures show moderate response rates that tend to improve over 12 to 24 months as neural recovery progresses, consistent with the concept of penile rehabilitation. The AUA guideline on erectile dysfunction recommends considering early PDE5 inhibitor use after nerve-sparing radical prostatectomy specifically to preserve long-term erectile function. [9]

Dose Escalation Over Time

A subset of long-term sildenafil users require dose escalation from 50 mg to 100 mg over the first 12 to 18 months, primarily in men with progressive vascular disease. That escalation need does not represent true tachyphylaxis but rather underlying disease progression reducing the endothelial signal sildenafil depends on. [10]

Long-Term Durability of Alprostadil: Caverject vs MUSE

Alprostadil's pharmacologic efficacy is extremely consistent because the mechanism does not depend on a deteriorating upstream pathway. The durability problem is behavioral and mechanical, not pharmacologic. [11]

Caverject (Intracavernosal Injection) Retention

Real-world retention data for Caverject are sobering despite its high acute efficacy. A prospective follow-up study of 178 men initiating Caverject found that 68% were still using it at 6 months, 55% at 12 months, and approximately 40% at 24 months. [12] Dropout reasons were injection anxiety (41%), penile pain (24%), partner objection (18%), and inadequate response (17%). Among men who continued past the 6-month mark, satisfaction remained high and dose requirements were stable, confirming true pharmacologic durability for those who persist.

Penile fibrosis is a long-term risk. A 5-year follow-up of Caverject users found palpable penile plaques in approximately 12% of men at 5 years, with an annual incidence of roughly 2 to 3% per year of use. [13] Most fibrosis is mild and does not require stopping therapy, but it warrants annual palpation of the corpora at follow-up visits.

MUSE (Intraurethral Pellet) Retention

MUSE data show consistently higher dropout rates than either oral sildenafil or Caverject. A North American registry of MUSE users followed 1,511 men and found that only 49.6% of initial responders continued using the pellet at 3 months. [14] At 12 months, fewer than 30% of the original cohort remained active users. Reasons included urethral burning (64% of dropouts), inadequate rigidity (29%), and hypotension (7%). The 1000 mcg dose produced the strongest erections but also the highest rate of urethral discomfort and systemic side effects.

Combined Use: When Clinicians Layer Both

A growing body of evidence supports combining low-dose sildenafil with low-dose Caverject or MUSE in men who respond partially to either agent alone. A small randomized crossover trial published in the Journal of Urology found that sildenafil 50 mg plus intracavernosal alprostadil 5 mcg produced successful intercourse in 92% of attempts versus 74% for sildenafil 100 mg alone in men with moderate ED, using lower alprostadil doses that reduced pain rates. [15] This combination approach is not FDA-labeled but appears in specialist guidelines as an option for refractory ED.

Head-to-Head Comparison: Sildenafil vs Alprostadil in the Same Patients

Direct Comparative Data

Head-to-head randomized trials comparing sildenafil and alprostadil are limited. A Cochrane systematic review on PDE5 inhibitors confirmed sildenafil's superiority over MUSE on patient preference and global satisfaction when both had been tried, with 79% of patients preferring oral sildenafil. [16] Direct comparisons with Caverject are complicated by the fact that many Caverject users have already failed sildenafil, making naive comparisons biased toward alprostadil appearing less effective.

A 2003 crossover study in men who were sildenafil-naive enrolled 50 participants to receive both sildenafil 100 mg and Caverject 10 mcg in randomized sequence. Successful intercourse rates were 74% for sildenafil and 88% for Caverject, with Caverject achieving greater penile rigidity as measured by RigiScan monitoring. Patient preference still favored sildenafil in 66% of cases, primarily due to route of administration. [17]

Side Effect Profiles and Their Effect on Long-Term Adherence

Sildenafil's side effects include headache (16%), flushing (11%), and visual disturbance (3%), all of which are typically mild and transient. [4] These effects do not worsen with prolonged use and rarely drive discontinuation after the first few doses. Caverject's main long-term concern is penile pain, reported as moderate or severe in 11% of users in the Linet trial. [5] MUSE's urethral burning affects a larger proportion of users, with rates between 30 and 65% across studies. [14]

The FDA label for alprostadil injection specifically warns of prolonged erection (duration exceeding 4 hours) and priapism, with an incidence of approximately 0.4% per injection in clinical trials. [18] Men must be counseled at initiation that an erection lasting more than 4 hours requires emergency treatment. Sildenafil-associated priapism is considerably rarer, estimated at less than 0.001% of exposures. [19]

Switching From Sildenafil to Alprostadil: Clinical Decision Points

When to Consider Switching

Most urologists and sexual medicine specialists reserve alprostadil for men who have failed an adequate trial of a PDE5 inhibitor. An adequate trial means at least 6 to 8 attempts at the maximum tolerated dose (sildenafil 100 mg) with proper timing (1 hour before activity, low-fat meal, adequate sexual stimulation). Men who skip that optimization step and are switched prematurely represent a common cause of unnecessary escalation. [9]

Clear indications for switching to alprostadil include:

• Failed response after an optimized sildenafil trial at 100 mg
• Contraindication to nitrates that make PDE5 inhibitors unsafe (sildenafil is absolutely contraindicated with organic nitrates) [18]
• Post-prostatectomy without nerve-sparing, where sildenafil response rates may be below 20%
• Severe arterial insufficiency with preserved venous function, where alprostadil's direct vasodilatory effect is more effective

Sequencing: Caverject Before MUSE, or MUSE Before Caverject?

Many clinicians trial MUSE before Caverject given the lower barrier to needle use. The evidence, though, suggests this sequencing underdelivers. MUSE's efficacy ceiling is lower, and men who fail MUSE and then try Caverject may attribute the overall experience to alprostadil class failure rather than route failure. A pragmatic approach is to offer Caverject as the primary injectable option with MUSE reserved for men with specific contraindications to injection (bleeding disorders, anticoagulant therapy) or confirmed needle phobia after counseling. [20]

Combination Therapy as an Alternative to Full Switching

Before a complete switch, partial responders to sildenafil may benefit from adding MUSE 250 to 500 mcg as a topical augment rather than moving to full Caverject injections. This combination is not approved by the FDA as a fixed regimen but has appeared in prospective case series with acceptable safety, provided blood pressure is monitored at initiation given additive hypotensive risk. [15]

Patient Selection: Who Does Best With Each Option

Sildenafil Is the Correct First-Line Agent For

Oral sildenafil remains first-line for most men with ED of psychogenic, mild vascular, or mixed etiology. The 2018 AUA Guideline on Erectile Dysfunction assigns PDE5 inhibitors a strong recommendation for first-line use, citing their oral route, favorable safety profile, and broad evidence base. [9] Men with preserved endothelial function, mild-to-moderate ED, and no history of prostate cancer treatment are the best long-term responders.

Alprostadil Is Preferred For

Alprostadil produces clinically meaningful responses in men for whom sildenafil rarely works. That group includes men with complete sacral spinal cord injury (sildenafil requires at least partial sacral reflex arc integrity), bilateral non-nerve-sparing prostatectomy, and severe corporal veno-occlusive dysfunction. A prospective study of 98 spinal cord injury patients found that intracavernosal alprostadil produced successful erections in 74% of complete sacral lesion patients versus 9% for sildenafil. [21]

Men on nitrate therapy for angina represent a separate sildenafil contraindication group. PDE5 inhibitors are absolutely contraindicated with any organic nitrate due to risk of severe hypotension, per the FDA label. [18] For these men, alprostadil is often the only pharmacologic option short of a penile prosthesis.

Practical Dosing and Titration for Long-Term Use

Sildenafil Dosing for Sustained Use

The approved sildenafil dose range is 25 to 100 mg taken 30 to 60 minutes before sexual activity. Most men begin at 50 mg. A 2019 real-world analysis of 4,200 sildenafil prescriptions found that 58% of men eventually required up-titration to 100 mg for sustained response over 24 months, particularly those with diabetes or hypertension. [22] Daily low-dose sildenafil 25 mg has been studied as a penile rehabilitation strategy and may support endothelial health even on days without sexual activity. [9]

Caverject Dosing Protocol

The FDA label for Caverject (alprostadil for injection) specifies starting doses of 1.25 to 2.5 mcg for neurogenic ED and 2.5 to 5 mcg for vasculogenic or psychogenic ED, with uptitration in-office until a satisfactory erection is achieved, not exceeding 40 mcg per injection. [18] The maximum recommended frequency is once daily or three times per week. Men should not self-inject until they have demonstrated correct technique under clinician supervision, including recognition of the signs of prolonged erection.

MUSE Dosing Protocol

MUSE is available in doses of 125, 250, 500, and 1000 mcg. The FDA label recommends starting at 250 mcg with in-office testing of the first dose to monitor for hypotension and urethral discomfort. [18] The 1000 mcg pellet produces the highest efficacy but also the highest rates of urethral pain and systemic vasodilation. A urethral warming and void-before-use protocol reduces systemic absorption variability and may marginally improve tolerability. [3]

Cost, Access, and Real-World Adherence Drivers

Generic sildenafil became widely available after Pfizer's U.S. Patent expiration in 2017. Prices dropped from more than $60 per tablet to under $1 per tablet for some generic formulations by 2020. [22] That price reduction substantially improved long-term adherence by eliminating the cost barrier that had previously driven many men to use sildenafil less frequently than prescribed, reducing effectiveness.

Caverject remains brand-name only in many markets. At 2024 U.S. Pharmacy prices, a single Caverject Impulse 20 mcg vial costs approximately $70 to 90 without insurance, making frequent use costly. MUSE pellets range from $50 to 80 per unit at retail. These cost differentials meaningfully affect long-term adherence: men using branded alprostadil average fewer doses per month than sildenafil users, according to pharmacy benefit data. [23]

Men who access sildenafil through telehealth and generic programs, where 90-day supplies cost $30 to 60, show higher 12-month continuation rates than those using retail pharmacies, a finding replicated across multiple insurance claim analyses. [23]

Summary Comparison Table

| Feature | Sildenafil (Viagra) | Caverject (Injection) | MUSE (Pellet) | |---|---|---|---| | Route | Oral | Intracavernosal injection | Intraurethral pellet | | Onset | 30 to 60 min | 5 to 20 min | 5 to 10 min | | Key trial success rate | 69% (Goldstein 1998) | 94% (Linet 1996) | 43 to 65% | | 12-month continued use | ~70 to 75% | ~55% | <30% | | Requires NO pathway | Yes | No | No | | Priapism risk | <0.001% | ~0.4%/injection | ~0.1% | | Penile fibrosis risk | None | ~2 to 3%/year | Rare | | FDA approval year | 1998 | 1995 | 1997 |