PSA: Drugs That Distort This Test

Updated May 25, 2026

Image: HealthRX.com clinical illustration

Prostate-specific antigen (PSA) is a serum blood test, not a diagnostic test for prostate cancer by itself. It measures a protein made by prostate epithelial cells, and several common medication classes change that number without changing cancer risk. The clinically important classes are 5-alpha reductase inhibitors (finasteride, dutasteride), which lower PSA; exogenous testosterone, which raises it; and a group of drugs, statins, NSAIDs, and thiazide diuretics, associated with smaller reductions. A PSA result interpreted without knowing the patient's medication list can send a falsely reassuring number past a clinician, or trigger a biopsy referral that reflects a drug effect rather than disease.

This article distinguishes what is well established (the direction and rough scale of these effects, reflected in prescribing information and specialty society guidance), what is plausible but less certain (precise correction formulas, combined-drug effects), and what individual readers should not attempt on their own (self-adjusting or stopping medication before a test, or interpreting a corrected PSA value without a clinician).

The direct answer

5-alpha reductase inhibitors are the largest and most predictable PSA distorter, lowering PSA by roughly half after sustained use, which is why prescribing information for these drugs instructs clinicians to establish a new baseline PSA after treatment starts and to treat any rise from that suppressed baseline as a signal, not just an absolute PSA value above 4.0 ng/mL. Testosterone therapy produces a smaller, more variable rise, most guidance describes it in fractions of a nanogram per milliliter over the first year. Statins, NSAIDs, and thiazides are associated with modest downward shifts that matter mainly when a result sits near a decision threshold. None of these effects should be assumed to apply at a precise, individual percentage without checking current prescribing information, because published effect sizes vary across studies and populations.

Evidence boundary: what this page can and cannot tell you

Established: 5-alpha reductase inhibitors lower PSA, and current prescribing information for finasteride and dutasteride instructs prescribers to obtain a new baseline PSA after the drug reaches steady state. Testosterone therapy raises PSA in most men who start it, and the Endocrine Society's 2018 clinical practice guideline on testosterone therapy in hypogonadal men requires PSA monitoring at baseline, at 3 to 6 months, and annually thereafter. The 2018 USPSTF recommendation on prostate cancer screening frames PSA-based screening in men 55 to 69 as an individual, shared decision rather than routine testing, precisely because PSA is an imperfect signal even before drug effects are added.

Plausible but not settled with precision: The commonly cited figures for how much statins, NSAIDs, or thiazide diuretics lower PSA (ranges in the single digits to low teens as a percentage) come from observational cohort and pharmacoepidemiologic studies rather than regulatory labeling, and the exact magnitude varies by drug potency, dose, and population studied. Readers and clinicians should treat any specific percentage from a single study as illustrative rather than a number to plug into a calculation.

Not established: There is no validated, universally endorsed formula for correcting PSA in a patient taking several PSA-altering drugs at once (for example, testosterone plus a 5-alpha reductase inhibitor, or a statin plus a thiazide). Multiplying and adding individual effect estimates from separate studies is a reasoning shortcut, not a validated clinical algorithm, and no major guideline currently endorses it for individual clinical decisions.

5-alpha reductase inhibitors: the largest known distortion

Finasteride and dutasteride block the conversion of testosterone to dihydrotestosterone (DHT). Because DHT drives prostate epithelial growth, PSA production falls as the gland shrinks under treatment. This is an on-label, well-documented pharmacologic effect, not an off-label or incidental one, and it is large enough that urology guidance treats it as a required adjustment factor rather than a footnote.

The commonly cited approximation is that PSA falls by around half after roughly six months of continuous therapy, a figure that traces to large finasteride and dutasteride prostate-cancer-prevention trials. Because the exact percentage varies somewhat by trial and by individual patient, clinicians typically respond to the trend from a post-treatment baseline rather than treating the "times two" correction as an exact number. A PSA velocity increase from that suppressed baseline, rather than a single absolute value, is the signal urology guidance treats as concerning, even when the raw PSA is still under 4.0 ng/mL.

An added complication: finasteride use in the largest prevention trials was associated with a higher proportion of higher-grade tumors detected among the cancers that were found, which some researchers have interpreted as a possible detection artifact (higher-grade tumors may produce less PSA per unit of tumor under DHT suppression) rather than a true increase in aggressive cancer caused by the drug. This nuance means a "corrected" PSA that looks reassuring should not be treated as ruling out higher-grade disease. If you take a 5-alpha reductase inhibitor, ask your clinician to document your pre-treatment PSA and your post-stabilization (roughly 6-month) PSA as two distinct reference points, and to describe how future results will be interpreted relative to the suppressed baseline rather than the general population reference range.

PSA interpretation decision framework, by medication class

Use this as a conversation guide with a clinician, not as a self-correction tool. It organizes the few facts that should change what happens next.

Medication class	Direction of PSA effect	Rough scale (verify current specifics)	What should change in interpretation	Who should not skip a workup anyway
5-alpha reductase inhibitors (finasteride, dutasteride)	Lowers PSA	Roughly halved after sustained use	Establish a new post-treatment baseline; watch trend/velocity, not the raw number against the standard 4.0 ng/mL cutoff	Any rise from the suppressed baseline, even if the absolute value stays low
Testosterone therapy	Raises PSA	Typically a fraction of a ng/mL rise over the first year, wide individual variation	Expect an early rise; guideline-based monitoring at baseline, 3 to 6 months, and annually is the standard response, not automatic discontinuation	A rise that clearly exceeds the guideline's stated threshold, or any new abnormal exam finding
Antiandrogens / GnRH agonists or antagonists (for prostate cancer treatment)	Lowers PSA sharply	Often to very low or undetectable levels within months	PSA becomes a treatment-response marker, not a screening test, in this context	Not applicable to screening decisions; interpretation belongs to the treating oncology/urology team
Statins	May lower PSA modestly	Small, single-digit-to-low-double-digit percentage in some cohort studies	Note statin use when a result sits close to a decision threshold; do not assume a normal-looking PSA rules out disease in a statin user with other risk factors	Family history, abnormal exam, or urinary symptoms
NSAIDs, including regular high-dose use	May lower PSA modestly	Small percentage reduction reported in observational studies, more heterogeneous than statin data	A recent unexplained PSA drop coinciding with starting daily NSAID use may be pharmacologic; discuss whether a retest after a short washout is appropriate	Any new symptoms independent of the PSA trend
Thiazide diuretics	May lower PSA modestly	Small percentage reduction reported in some population studies	Relevant mainly at borderline values close to a biopsy threshold	Long-term thiazide users with a borderline PSA and other risk factors deserve the same scrutiny as anyone else
Recent ejaculation, prostate biopsy, or urinary tract infection	Raises PSA temporarily	Can be a marked, short-lived increase	Reschedule testing rather than reacting to an isolated spike	Persistent elevation after the expected recovery window
Significant weight loss (including GLP-1 therapy) or obesity	Obesity dilutes PSA; weight loss can unmask a higher underlying value	Direction is physiologically plausible (blood volume changes PSA concentration); exact magnitude is not well pinned down	A PSA rise coinciding with major weight loss is not automatically alarming, but it should be discussed rather than assumed benign	Any rise paired with a new abnormal exam finding

The one-line decision rule underneath this table: trend and context beat a single number. A PSA reading only becomes actionable once you know what medications the patient is taking, what the pre-treatment baseline was, and whether the current value represents a change from that person's own recent history.

Testosterone replacement therapy: what the guideline actually asks for

As testosterone converts locally to DHT, it drives prostate epithelial growth, which commonly results in elevated PSA readings during the initial year of treatment. According to The Endocrine Society's 2018 clinical practice guideline for testosterone replacement in hypogonadal men, PSA measurement should occur at baseline before initiating therapy, at 3 to 6 months into treatment, then yearly. Urology referral is warranted if PSA increases notably from baseline in the first treatment year or reaches the absolute level designated in the guideline. Readers should consult the published guideline directly for precise PSA numeric thresholds rather than relying on values stated here, as these recommendations may be revised and indirect citation can introduce inaccuracy.

Men who start testosterone therapy with a baseline PSA already in the upper-normal range are more likely to cross a conventional screening cutoff purely from the expected pharmacologic rise. Reviews of testosterone trial data have generally not found that testosterone therapy itself increases prostate cancer incidence, even though it reliably raises PSA, which is why specialty guidance frames an expected PSA rise as a monitoring trigger rather than an automatic reason to stop treatment. Men taking both testosterone and a 5-alpha reductase inhibitor at the same time (a combination sometimes used in benign prostatic hyperplasia management) are pulling PSA in opposite directions, and no validated single correction formula exists for that combination; serial measurement against each individual's own post-stabilization baseline is the more defensible approach.

Statins, NSAIDs, and thiazide diuretics: smaller but real shifts

Statins, regular NSAID or aspirin use, and thiazide diuretics have each been associated with modest reductions in PSA in observational and cohort studies, generally described as single-digit to low-double-digit percentage effects. The proposed mechanisms differ (anti-inflammatory action, effects on androgen synthesis, and plasma volume expansion/hemodilution, respectively), and none of these three drug classes has label language instructing PSA correction the way 5-alpha reductase inhibitors do. For most men the absolute PSA change from any one of these drugs is small, often a few tenths of a nanogram per milliliter, but for a patient whose true PSA sits close to a biopsy threshold, that shift can matter. No major guideline currently recommends a numeric correction factor for statin, NSAID, or thiazide use; the practical step is for clinicians to record medication use alongside the PSA result so a borderline value can be interpreted in context rather than treated as unambiguous.

Other medications and situations that change PSA

Antiandrogens (bicalutamide, enzalutamide, apalutamide) and GnRH agonists/antagonists (leuprolide, degarelix). These suppress androgen signaling or induce medical castration and are used in the treatment of prostate cancer itself, not for screening purposes. In that treatment context, PSA becomes a marker of response to therapy, and comparing it to general screening reference ranges is not meaningful.

Saw palmetto and other prostate supplements. A randomized trial found no significant PSA change from standard-dose saw palmetto. Unregulated combination supplements marketed for "prostate health" are a separate concern, because some contain undisclosed active ingredients, and their effect on PSA cannot be predicted from the saw palmetto evidence alone.

Obesity and significant weight loss (including GLP-1 receptor agonist therapy). Higher body mass is associated with plasma volume expansion, which dilutes PSA concentration independent of prostate biology. The reverse is biologically plausible during rapid, significant weight loss: PSA may appear to rise as blood volume normalizes, without any underlying change in the prostate. This is a physiologically reasonable mechanism, but a precise, quantified relationship between weight change and PSA change has not been pinned down here and should not be treated as a fixed correction.

Recent ejaculation, cycling, biopsy, or urinary tract infection. Each of these can transiently raise PSA, sometimes substantially in the case of infection or recent biopsy. These are not medication effects, but they are common, avoidable sources of a misleading single reading.

Practical steps before a PSA test

Abstain from ejaculation in the day or two before the blood draw, since it can transiently raise PSA.
Avoid prolonged cycling or other perineal pressure in the 48 hours before the draw.
Do not schedule a screening PSA within about six weeks of a documented urinary tract infection or a prostate biopsy; both can cause a large, temporary spike.
Tell the ordering clinician about every prescription medication and supplement, including dose and how long you have taken it.
Use the same laboratory for repeat measurements when possible; assay differences between labs can matter at borderline values.
If you take a 5-alpha reductase inhibitor, ask that your chart clearly note both your pre-treatment PSA and your post-stabilization baseline, so future results are compared to the right reference point.

None of these steps replace a clinician's judgment, and none should be used to talk yourself out of a recommended follow-up test.

When to seek care rather than wait for the next scheduled test

A single elevated or unusually low PSA is rarely an emergency. Seek prompt evaluation rather than waiting if you have new urinary symptoms (difficulty urinating, blood in urine or semen, pelvic or back pain) alongside an abnormal PSA, or if a clinician has told you your PSA rise meets a threshold that warrants same-window follow-up rather than routine annual testing. If you are on testosterone therapy or a 5-alpha reductase inhibitor and notice a PSA change your clinician has not yet addressed, raise it directly rather than assuming the drug explains it.

Frequently asked questions

What is a normal PSA level?

A PSA between 0 and 4.0 ng/mL is commonly used as a general reference range, with lower age-adjusted thresholds sometimes used for younger men and higher ones for older men. These reference ranges assume no interfering medication, recent ejaculation, infection, or biopsy, all of which can shift the result.

What does a high PSA mean?

An elevated PSA can reflect prostate cancer, benign prostatic hyperplasia, prostatitis, recent ejaculation, a urinary infection, or a medication effect such as testosterone therapy. PSA alone cannot distinguish between these causes, which is why an abnormal result is typically followed by additional testing or clinical evaluation rather than an immediate biopsy.

Does finasteride hide prostate cancer on PSA tests?

Finasteride and dutasteride lower PSA substantially, which is why prescribing guidance asks clinicians to establish a new baseline after treatment starts and to watch for a rise from that baseline rather than relying on the general population reference range. There is also evidence suggesting higher-grade cancers may produce disproportionately less PSA under this suppression, so a reassuring corrected number does not eliminate the need for clinical follow-up when other risk factors are present.

How much does testosterone therapy raise PSA?

Most men see a modest PSA rise during the first year of testosterone therapy, and specialty guidance requires monitoring at baseline, 3 to 6 months, and annually, with defined thresholds for urological referral. The exact numeric threshold should be confirmed against the current Endocrine Society guideline rather than assumed from a single fixed figure.

Should I stop my medications before a PSA test?

Do not stop a prescribed medication without your clinician's guidance. Report all medications and supplements so the result can be interpreted correctly. Whether a short pause of an as-needed NSAID before testing is appropriate is a decision for your clinician, not something to do unilaterally.

Do statins lower PSA enough to miss cancer?

Statins are associated with a modest PSA reduction in observational studies. For most men this shift is too small to change a screening decision, but for someone whose PSA sits close to a biopsy threshold, it is a factor worth mentioning to the ordering clinician.

Can saw palmetto affect my PSA results?

A randomized trial found no significant PSA change with standard-dose saw palmetto. Unregulated combination prostate supplements are a separate concern, since undisclosed ingredients could affect PSA in ways the saw palmetto evidence does not cover.

Does weight loss raise PSA?

Obesity is associated with lower PSA through blood volume expansion (hemodilution), and significant weight loss could plausibly reverse that effect and raise the measured PSA without any true change in the prostate. The size of this effect has not been precisely established and should be discussed with a clinician rather than assumed.

Can a urinary tract infection raise PSA?

Yes. Bacterial prostatitis and urinary tract infections can raise PSA substantially. Screening PSA testing is typically deferred for several weeks after a documented infection resolves to avoid a false-positive result.

Sources referenced in this draft

USPSTF, Prostate Cancer Screening recommendation (2018): https://www.uspstf.org/recommendation/prostate-cancer-screening
Endocrine Society, Testosterone Therapy in Men With Hypogonadism clinical practice guideline (2018), Journal of Clinical Endocrinology & Metabolism: https://academic.oup.com/jcem/article/103/5/1715/4939465

The original draft of this article carried numbered citations to specific PubMed identifiers for individual trials and cohort studies (finasteride prevention trial data, dutasteride trial data, testosterone trial data, statin and NSAID cohort studies, thiazide studies, and others). Those identifiers are unverified against the underlying papers and have been removed from this draft rather than repeated with false precision. Before publication, an editor or reviewer with database access should confirm each specific numeric claim (percentage reductions, hazard ratios, sample sizes) against the primary literature or current FDA prescribing information and reinstate only the citations that check out.