Issue 08 looked at why raising the same vascular signal can produce very different results from one person to another. This paper follows that signal into a completely different part of the cell: the machinery that moves cholesterol.

Read Issue 08: When Testosterone Goes Up and Libido Goes Down →

4:04 with Dr. Gupta — the elegant mechanism, the human signal, and why the finding does not change a prescription today.

An unexpected effect of PDE5 inhibition made cancer cells less capable of spreading in experimental models. Whether it does the same in patients remains unknown.

Maybe. But if we are being precise—the kind of precision that matters before changing a medication—the strongest evidence so far comes from cancer cells and mice, not from treating cancer patients.

Viagra and Cialis are familiar examples of PDE5 inhibitors. We use them because they raise cyclic GMP, relax smooth muscle, and improve blood flow for erections. A new Cancer Research paper suggests that the same rise in cyclic GMP may have another, completely unexpected effect: it can interfere with the way cancer cells make cholesterol available for the work of spreading.

Medicine is humbling: a drug can be used for decades and still reveal biology we were not looking for. Good science stays open to surprises without pretending they have already answered the clinical question.

PDE5 normally breaks down cyclic GMP. Block PDE5, and cyclic GMP rises.

In these experiments, higher cyclic GMP interfered with NPC1, the protein that moves cholesterol out of the lysosome. It did not shut down cholesterol production. It trapped more cholesterol in storage, leaving less available for membranes, signaling, mitochondrial function, and migration.

That is what makes the arc so elegant: the mechanism predicted less movement, and the models showed reduced migration and metastatic burden.

Sildenafil was the principal drug, but the investigators also tested vardenafil and genetically reduced PDE5A. A statin strengthened the effect in some models by reducing new cholesterol synthesis through a different step.

Itraconazole provides a useful precedent: the antifungal can bind NPC1 directly and block lysosomal cholesterol export. It does not prove benefit from PDE5 inhibitors, but it shows the transporter is pharmacologically tractable.

This was an antimetastatic result in experimental systems—not evidence that Viagra prevents cancer or reduces metastasis in patients.

The same paper also looked backward through health records from 40,567 cancer patients aged 35–65. The researchers did not give sildenafil to treat cancer. They classified sildenafil and statin dispensations during the six months before cancer diagnosis, then examined five-year survival.

One or two sildenafil dispensations were not significantly associated with lower mortality: HR 0.92, 95% CI 0.77–1.09. Three or more dispensations were associated with lower mortality: HR 0.74, 95% CI 0.55–0.99. With statin exposure added, the corresponding estimates were HR 0.81, 95% CI 0.67–0.91, for one or two sildenafil dispensations and HR 0.68, 95% CI 0.50–0.91, for three or more, compared with neither exposure.

This was a retrospective population analysis, not a treatment trial. Dispensing frequency is not a verified dose or proof of adherence. The database also lacked cancer stage and cancer treatment—both major determinants of survival.

This is the point where medicine requires restraint. Association is not causation. The analysis found a signal worth pursuing; it did not show that sildenafil prolonged life.

An older Swedish registry asked a different question. Among men with prior benign colorectal neoplasms, colorectal-cancer incidence was 2.64 per 1,000 person-years in PDE5-inhibitor users versus 4.46 in nonusers: adjusted HR 0.65, 95% CI 0.49–0.85.

The association strengthened across cumulative exposure, but a trend does not make a study randomized. Healthy-user effects, surveillance differences, residual confounding, and time-related bias can remain. It raises a prevention hypothesis; it does not establish prevention.

Cancer is not one disease. In experimental BRAF-mutant melanoma, PDE5A suppression and pharmacologic inhibition increased invasion through a different cyclic-GMP, calcium, and contractility pathway—not the NPC1 mechanism.

Human studies have reported a small association with melanoma but have not established causation. The counterexample matters: one favorable mechanism should never become a universal cancer story.

There is a real possibility that erection medicines such as Viagra and Cialis could do more than improve blood flow. In experimental models, PDE5 inhibition also made cancer cells less effective at spreading. The mechanism is plausible, the experimental result is compelling, and the population data are interesting.

But possibility is the correct word.

I would not tell someone to take a PDE5 inhibitor because it prevents cancer. I would not start one, remain on one, switch from sildenafil to tadalafil, or increase a dose for an oncologic reason. These medications were not given as cancer treatment in the human analyses, and the associations do not establish causation.

If you already take one, the reason you were prescribed it remains the reason to take it. Nothing in this issue is a reason to start, stop, switch, or change a dose without the clinician managing your care.

Read the full dispatch →

The only thing we can be sure of is uncertainty. Good science asks us to question what we think we know, stay open to surprises, and remain just as rigorous about what the evidence has not shown.

The fine print. This is general education, not a recommendation to start, stop, combine, or change any prescription medication on your own. Treatment decisions require an individual physician evaluation and ongoing clinical oversight. Kinetic Edge Health provides care in select states.

If an article like this raises a question about a medication you take, the next step is a physician review of the whole picture — not a change made on your own.

Short-form dispatches on longevity, metabolic health, and men's performance — from the founder of Kinetic Edge Health.

Physician-led telehealth medical practice. Licensed in select states. Content is educational and does not constitute medical advice. Peptides and compounded medications are prescription items requiring physician evaluation and ongoing clinical oversight.

Privacy · Terms · Telehealth Consent

kineticedgehealth.com · Cleveland, OH

© 2026 Kinetic Edge Health, Inc. · All rights reserved.