evipedia.ai
← Back

Pirfenidone to Treat Cancer

Evidence Review created on 09/07/2026 using AI4L / Opus 5

Also known as: Esbriet, Pirespa, Pirfenex, Etuary, Deskar, 5-methyl-1-phenylpyridin-2(1H)-one

Motivation

Pirfenidone is an oral medicine, licensed since the late 2000s, that slows the build-up of scar tissue in the lungs. It calms the signals that push connective-tissue cells to lay down collagen, and it damps inflammation and oxidative damage as well. Tumours recruit very similar connective-tissue cells to build the dense, stiff scaffold around them — a scaffold that shields them from immune attack and blunts the delivery of drugs. That overlap is why an old lung medicine is now being tested in cancer care.

Interest has grown along two separate tracks. One asks whether the drug can soften a tumour’s supporting tissue enough to make chemotherapy, radiation, and immune-based treatment work better. The other asks whether it can protect the lungs of people whose cancer treatment — surgery, radiation, or drugs — threatens to trigger or worsen scarring. The two tracks rest on very different bodies of evidence.

This review sets out what has been measured in people, what is still confined to cells and animals, what the medicine costs in side effects, and where the boundary between those tracks currently sits.

Benefits - Risks - Protocol - Conclusion

High-level sources that examine pirfenidone’s activity against tumours and its behaviour in people receiving cancer treatment.

Of the six priority expert platforms, only Life Extension mentions pirfenidone at all — a single naming of the drug inside its idiopathic pulmonary fibrosis protocol, with nothing on cancer — so none of the six supplied content that meets the relevance bar for this review. The drug is a prescription antifibrotic with no consumer or longevity following, and these platforms cover supplements, lifestyle interventions and mainstream longevity pharmacology rather than niche repurposed respiratory medicines.

Grokipedia

Pirfenidone

Covers the drug end to end — approvals, pharmacology, adverse effects and interactions — and its research section summarises the radiation-induced lung injury evidence in cancer patients that this review examines.

Examine

No Examine article exists for pirfenidone.

Pirfenidone is a prescription medication rather than a dietary supplement, and Examine.com does not typically cover prescription medications.

ConsumerLab

No ConsumerLab article exists for pirfenidone.

Pirfenidone is a prescription medication and not a tested consumer supplement, and ConsumerLab does not typically cover prescription medications.

Systematic Reviews

Pooled evidence on pirfenidone relevant to people with cancer, covering both the claimed protective effect and the tolerability cost, most of it drawn from idiopathic pulmonary fibrosis (IPF, progressive scarring of the lungs with no identified cause), the drug’s licensed setting.

Both sides of the trade-off are represented: the claimed effect by Srivali et al., Srivali & De Giacomi and Petnak et al., the principal risk by Kou et al. and Wu et al. No systematic review or meta-analysis exists for the anti-tumour treatment claim itself, and that side of the ledger is unrepresented in the pooled literature.

Mechanism of Action

Pirfenidone is a small synthetic pyridinone with no single high-affinity receptor. Its dominant action is suppression of transforming growth factor beta 1 (TGF-β1, the master signal that tells connective-tissue cells to make scar tissue), which reduces the conversion of fibroblasts into collagen-secreting myofibroblasts and lowers deposition of extracellular matrix (ECM, the protein scaffold between cells). It also lowers tumour necrosis factor alpha and interleukin-1β (inflammatory messenger proteins) and scavenges reactive oxygen species.

In tumours the same fibroblast programme builds the desmoplastic stroma (the stiff, collagen-rich shell around a tumour). Cancer-associated fibroblasts (CAFs, connective-tissue cells recruited by tumours) secrete chemokines that draw in suppressive immune cells and keep killer T cells out, raise pressure inside the tumour, and limit drug penetration. Pirfenidone reduces CAF activation, suppresses TGF-β1/Smad signalling (Smads relay that signal inside the cell) and the epithelial-mesenchymal transition (EMT, the shift that lets tumour cells detach and migrate), and in breast models blocks Hedgehog/GLI1 signalling (a developmental pathway switched back on in some tumours).

Two mechanistic readings compete. One holds that softening the stroma improves drug delivery and immune access. The other, drawn from pancreatic work on other stroma-depleting agents, holds that the fibrotic shell also restrains invasion, so stripping it may speed spread.

Pharmacologically: terminal half-life roughly 2.4 hours; largely oxidised by the liver enzyme CYP1A2 (which also clears caffeine) to 5-carboxy-pirfenidone, with minor input from the related liver enzymes CYP2C9, CYP2C19, CYP2D6 and CYP2E1; elimination is mainly renal as that metabolite; distribution is broad rather than tissue-selective.

Historical Context & Evolution

Pirfenidone was synthesised in the 1970s by Solomon Margolin and first pursued as a fever reducer, painkiller and anti-parasitic agent. Its antifibrotic property surfaced only later, when animal work showed it blunted lung scarring caused by bleomycin (a chemotherapy drug used experimentally to injure the lung). Through the 1990s it was supplied compassionately to people with progressive lung scarring on the strength of open-label series.

Registration was uneven. Japan approved it in 2008 and Europe in 2011. The U.S. Food and Drug Administration (FDA) declined approval in 2010 because one of two twin phase 3 trials missed its lung-function endpoint, and cleared it only in 2014 after a third trial reproduced the effect. Those registration trials were designed, funded and co-authored by the manufacturer InterMune, later acquired by Roche — a direct commercial interest that shaped which questions were asked and which were left alone.

Oncology interest arose from two observations: scarred lungs carry a markedly raised lung cancer risk, and the fibroblasts inside tumours behave much like those in scarring tissue. The earliest human tumour tests were in neurofibromatosis type 1 (an inherited condition causing nerve-sheath tumours). An adult open-label study reported tumour volume falling at least 15% in four of twenty-four participants, while a later paediatric study with a comparison group found no objective responses and no gain in time to progression. Neither result closed the question; together they narrowed it to combination and lung-protection settings, where testing has since concentrated.

Expected Benefits

High 🟩 🟩 🟩

No benefit reaches High: every cancer-relevant human outcome rests on a single randomised controlled trial (RCT, a study in which participants are allocated to treatments by chance), a single-arm phase 2 study, or pooled observational cohorts — none has been reproduced in a second controlled trial.

Medium 🟩 🟩

Recovery of Lung Gas Transfer After Radiation-Induced Lung Injury

Thoracic radiotherapy scars healthy lung, and no medicine is standard treatment for it. In a ten-centre randomised phase 2 trial of 134 patients, adding pirfenidone to steroids improved gas-transfer capacity (how much oxygen crosses from air into blood) at 24 weeks, as reported by Hou and colleagues. A retrospective series of 33 lung cancer patients found radiographic improvement in 79%, including four with overlapping immune-treatment pneumonitis (lung inflammation caused by immunotherapy). The trial was open-label and short.

Magnitude: Gas transfer rose 8.0% from baseline on pirfenidone versus a 2.4% fall on steroids alone — a least-squares mean difference (the between-group gap after statistical adjustment) of 10.4% (95% confidence interval, the range in which the true value most likely lies, 4.3–16.5; p=0.0010, where p is the probability such a difference arose by chance) at 24 weeks.

Prevention of Fibrosis Flares During Cancer Treatment

People with scarred lungs face a 4.4–20% risk of a sudden, often fatal fibrosis flare after lung cancer surgery. A meta-analysis of four cohorts covering 261 patients found perioperative antifibrotic therapy cut both flares and early death, and a single-arm phase 2 trial in 39 surgical patients met its endpoint against a pre-specified threshold. A small retrospective cohort extended the pattern to chemotherapy and checkpoint immunotherapy (treatment that releases the brakes on immune cells). All comparative data are observational.

Magnitude: Flare risk ratio 0.31 (RR, the risk in treated patients divided by the risk in untreated patients; 95% confidence interval 0.13–0.70) and 90-day mortality risk ratio 0.19 (0.07–0.52); in the single-arm trial 94.9% remained flare-free versus an 80% null value (p=0.01).

Low 🟩

Lower Lung Cancer Incidence in Progressive Lung Scarring ⚠️ Conflicted

A nationwide claims analysis of 9,938 patients found fewer new lung cancers on pirfenidone, replicated only weakly in a cohort of 941. A published methodological critique attributes that pattern to index-date choice, surveillance and competing risks. This is prevention, not treatment, and the net reading is an unconfirmed signal.

Magnitude: 10.4 versus 27.9 cases per 1000 person-years after weighting; weighted adjusted hazard ratio 0.347 (HR, the relative rate at which an event occurs over time; 95% confidence interval 0.258–0.466), falling to 0.716 (0.517–0.991) in the clinical cohort.

Volume Reduction in Fibrous Nerve-Sheath Tumours ⚠️ Conflicted

The only human tumour-shrinkage data come from neurofibromatosis type 1. An open-label adult study reported volume falls of at least 15% in a minority; a larger paediatric study benchmarked against a placebo group found none. The net reading is no reliable tumour-shrinking effect.

Magnitude: Four of 24 adults had tumour volume fall at least 15%, 17 were stable and three progressed; the paediatric study recorded no objective responses in 36 participants and a median time to progression of 13.2 versus 10.6 months (p=0.92).

Speculative 🟨

Sensitisation of Tumours to Chemotherapy and Radiotherapy

In cell and mouse models the drug amplifies killing by standard chemotherapy drugs, reported by Mediavilla-Varela and Branco. No human efficacy data exist; the basis is preclinical only.

Restoration of Immune Cell Entry Into Tumours

Reversing the dense stroma lets killer T cells back into pancreatic and breast tumours in mice, shown by Zhang and Mpekris. Evidence is mechanistic and animal-based, with no human readout.

Suppression of Metastatic Spread

The drug curbed migration, invasion and lung metastasis in triple-negative breast models by blocking transforming growth factor beta signalling, reported by Luo. Basis is animal and cell work; no human endpoint exists.

Benefit-Modifying Factors

  • Smoking and CYP1A2 activity: Cigarette smoke strongly induces CYP1A2 (the liver enzyme that clears most pirfenidone and caffeine), roughly halving drug exposure. Current smokers — a large fraction of any lung cancer population — may derive markedly less benefit at a standard dose.

  • CYP1A2 and MUC5B variants: Fast-metaboliser CYP1A2 variants lower drug levels; the MUC5B promoter variant (a mucin gene change that raises lung scarring risk) marks the fibrotic phenotype in which the protective benefits were demonstrated, and is absent in most other tumour settings.

  • Baseline gas transfer and lung volume: Benefit in the radiation-injury trial was measured as recovery of gas-transfer capacity from an already impaired baseline. Those with near-normal baseline lung function have far less room to improve and correspondingly less to gain.

  • Tumour stromal density: The mechanism requires abundant tumour-supporting fibroblasts. Stroma-rich tumours — pancreatic, some breast and lung types — offer a plausible target; stroma-poor tumours such as most blood cancers offer none.

  • Sex: Women have on average lower CYP1A2 activity and therefore higher drug exposure at the same dose, which may shift both effect and side-effect burden upward. Fibrosis cohorts were roughly three-quarters male, so female data are comparatively thin.

  • Pre-existing lung scarring: Every human benefit shown so far occurred in people whose lungs were already scarred or radiation-injured. Without that substrate the protective rationale disappears and only the unproven anti-tumour rationale remains.

  • Age: Fibrosis cohorts averaged near 70 years, so older adults are the best-characterised group. Drug exposure rises modestly with age, and weight loss and fatigue are less well tolerated at the older end of the range.

Potential Risks & Side Effects

High 🟥 🟥 🟥

Gastrointestinal Intolerance

Nausea, diarrhoea, dyspepsia (upper abdominal discomfort and heartburn) and vomiting are the signature adverse effects, arising from direct irritation of the gut lining and from nausea signalling in the brain rather than from the antifibrotic action. They cluster in the first three months, are usually mild to moderate, and are the leading reason people stop. In someone already nauseated by chemotherapy, this burden stacks rather than substitutes, as documented in the pooled five-trial analysis.

Magnitude: Across 1,299 trial participants, nausea 37.6%, diarrhoea 28.1%, dyspepsia 18.4%, vomiting 15.9%; in real-world pooled data 16.6% discontinued for adverse events and 56.4% reported any adverse event.

Photosensitivity and Rash

The drug and its metabolite absorb ultraviolet light, producing exaggerated sunburn and rash on exposed skin, typically within the first six months. This matters disproportionately in cancer care, where radiotherapy fields and several chemotherapy drugs already inflame skin, and where sun avoidance compounds vitamin D depletion. Reactions are dose-related and usually resolve on interruption, as reported in the CAPACITY trials.

Magnitude: Photosensitivity 12% versus 2% on placebo and rash 32% versus 12% in CAPACITY; rash 25.0% across the pooled five-trial cohort; grade 3 or worse rash in 3% of the pirfenidone arm of the radiation-injury trial versus none on steroids alone.

Elevated Liver Enzymes and Drug-Induced Liver Injury

Rises in alanine aminotransferase and aspartate aminotransferase (ALT and AST, blood enzymes that leak from injured liver cells) occur in a small but consistent minority and usually reverse on dose reduction or withdrawal, as quantified in the pooled safety analysis. Rarely, post-marketing reports describe serious drug-induced liver injury, including fatal cases. The concern in oncology is additive injury: many chemotherapy drugs, targeted drugs and checkpoint inhibitors are themselves toxic to the liver, and a shared signal can force the wrong drug to be stopped.

Magnitude: Alanine or aspartate aminotransferase above three times the upper limit of normal in 40 of 1,299 patients (3.1%), an adjusted incidence of 2.3 per 100 person-exposure-years.

Weight Loss and Reduced Appetite

Appetite suppression and weight loss are among the most consistent effects and are mechanistically linked to the gastrointestinal burden. In cancer this is the most consequential of the common adverse effects, because involuntary weight loss independently predicts treatment interruption and poorer outcome, and because cachexia (wasting of muscle and fat driven by the tumour itself) is already pulling in the same direction, as recorded in CAPACITY.

Magnitude: Anorexia (loss of appetite) in 11% of the higher-dose group versus 4% on placebo in CAPACITY; decreased weight ranks among the most frequently reported adverse reactions in the real-world pooled analysis, which reports no single pooled figure for the magnitude of loss.

Fatigue, Dizziness and Headache

Central effects appear early and persist at a lower rate. They are dose-related and reversible, but they compound the fatigue already imposed by radiotherapy and systemic cancer therapy and can be misattributed to disease progression, prompting unnecessary imaging or dose changes to the cancer regimen, as reported in CAPACITY and the nine-trial meta-analysis.

Magnitude: Dizziness in 18% of the higher-dose group versus 10% on placebo in CAPACITY; overall adverse-event relative risk (the treated group’s risk divided by the placebo group’s risk) 3.89 across nine randomised trials (p<0.01).

Medium 🟥 🟥

Serious Adverse Events During Cancer Treatment

In the only randomised trial run in a cancer population, adding the drug to steroids did not lower the serious adverse-event rate, and the small numerical excess was driven by skin toxicity rather than infection. Pneumonia, the main competing hazard, was less frequent on pirfenidone. No treatment-related deaths occurred. The trial was open-label, ran a median 9.2 months, and was not powered for rare harms, as reported by Hou and colleagues.

Magnitude: Serious adverse events in 12 of 67 (18%) on pirfenidone versus 11 of 67 (16%) on steroids alone; grade 3 or worse (severe or life-threatening) pneumonia 6% versus 12%; no treatment-related deaths in either arm.

Low 🟥

Severe Hypersensitivity Reactions

Angioedema and anaphylaxis (airway and facial swelling) sit alongside Stevens-Johnson syndrome, toxic epidermal necrolysis and drug reaction with eosinophilia and systemic symptoms (blistering, whole-body drug allergies) as post-marketing reports carrying labelled warnings. All are idiosyncratic, not dose-related, and demand permanent withdrawal, as described by de Perosanz-Lobo and Zhu.

Magnitude: Not quantified in available studies. Only isolated post-marketing case reports exist, so no incidence rate has been established in any trial or cohort.

No Demonstrated Survival Advantage in Cancer Cohorts

The only cohort comparing survival after a cancer diagnosis between antifibrotic-treated and untreated patients found no advantage, with unadjusted mean survival numerically shorter on treatment. The authors state the comparison cannot be adjusted and should not be over-read. It remains the sole direct human survival evidence, reported by Bertuccio.

Magnitude: Mean overall survival from cancer diagnosis 17.9 months with antifibrotic therapy versus 33.2 months without, in 61 patients; the study reports no adjusted estimate because censoring data were missing.

Speculative 🟨

Stromal Depletion Enabling Tumour Escape

Removing the fibrotic shell may release a restraint on invasion. In a mouse liver model the drug reduced scarring but left tumour incidence, size and number unchanged, reported by Seniutkin. Basis is animal and mechanistic.

Risk-Modifying Factors

  • CYP1A2 variants: Slow-metaboliser CYP1A2 genotypes raise drug exposure and with it the gastrointestinal, skin and liver burden. No genotype-guided dosing exists, so the practical marker remains observed tolerability at the starting dose.

  • Smoking status: Smoking induces CYP1A2 and roughly halves exposure, cutting side effects at the cost of effect. Abrupt smoking cessation during treatment can raise levels sharply and unmask toxicity that was previously absent.

  • Baseline liver enzymes and bilirubin: Pre-treatment alanine aminotransferase, aspartate aminotransferase and bilirubin above normal predict a higher chance of enzyme rises forcing dose reduction, and make attribution impossible when a liver-toxic cancer drug is running concurrently.

  • Baseline body weight and albumin: Low weight and low albumin at baseline mark the people in whom further appetite suppression is least affordable, and in whom weight-driven interruption of cancer therapy is most likely.

  • Sex: Women’s lower average CYP1A2 activity yields higher exposure and, in practice, a somewhat greater rate of nausea and photosensitivity at the same dose. Trial populations were male-predominant, so female-specific rates are less certain.

  • Pre-existing conditions: Impaired liver function raises exposure and enzyme-rise risk; impaired kidney function slows clearance of the active metabolite; inflammatory bowel disease and radiotherapy enteritis (radiation injury to the bowel lining) amplify the gastrointestinal burden.

  • Age: Exposure rises modestly in older adults, and both weight loss and dizziness carry greater consequence — falls, muscle loss, treatment interruption — at the older end of the target range than the raw incidence figures suggest.

Key Interactions & Contraindications

  • Strong CYP1A2 inhibitors (drugs that block the enzyme, e.g. fluvoxamine): Fluvoxamine, an antidepressant, roughly quadruples exposure. Absolute contraindication; consequence is severe gastrointestinal toxicity, rash and enzyme rises. Mitigation is substituting the interacting drug, not reducing pirfenidone.

  • Moderate CYP1A2 inhibitors (ciprofloxacin, enoxacin): Ciprofloxacin at 750 mg twice daily raises exposure roughly 80%. Caution; consequence is amplified nausea and photosensitivity. Mitigation is reducing pirfenidone to 1602 mg per day for the antibiotic course.

  • Combined multi-pathway inhibitors (fluvoxamine plus fluconazole, amiodarone, propafenone): Blocking CYP1A2 alongside other clearance routes compounds exposure. Caution; consequence is cumulative toxicity. Mitigation is avoiding concurrent use, or halving the dose with weekly symptom review.

  • CYP1A2 inducers (drugs that speed the enzyme up: rifampicin, omeprazole, carbamazepine): These lower exposure substantially. Caution; consequence is silent loss of effect rather than toxicity. Mitigation is a non-inducing alternative — famotidine in place of omeprazole, for example.

  • Over-the-counter medications (omeprazole, esomeprazole, ibuprofen, naproxen): The acid suppressants induce the enzyme; the anti-inflammatories irritate the gut. Caution; consequence is loss of effect and compounded nausea. Mitigation is switching the acid suppressant and taking any anti-inflammatory with food.

  • Photosensitising drugs (doxycycline, levofloxacin, hydrochlorothiazide, amiodarone): These add to ultraviolet sensitivity, as do several chemotherapy drugs. Caution; consequence is severe phototoxic burn. Mitigation is sun protection factor 50 or higher sunscreen plus covering clothing.

  • Caffeine and melatonin: Both are cleared by CYP1A2 and compete with pirfenidone for it. Caution; consequence is exaggerated tremor, palpitations or morning drowsiness. Mitigation is halving caffeine intake at initiation and reassessing after two weeks.

  • Cruciferous (broccoli-family) and enzyme-inducing supplements: Indole-3-carbinol, diindolylmethane and St. John’s wort induce CYP1A2 and lower drug levels. Caution; consequence is loss of effect. Mitigation is avoiding concentrated extracts, though ordinary dietary vegetable intake is not a concern.

  • Supplements with additive antifibrotic action: N-acetylcysteine, curcumin and silymarin act on overlapping oxidative and transforming growth factor beta pathways. Monitor; combined use has been formally studied only for N-acetylcysteine, where added benefit was not established and gastrointestinal burden rose.

  • Other interventions: Concurrent thoracic radiotherapy and checkpoint immunotherapy overlap with pirfenidone on skin and lung toxicity. Monitor; consequence is difficulty attributing a rash or pneumonitis to the right agent. Mitigation is staggering initiation by at least two weeks where the schedule allows.

Populations who should avoid Pirfenidone:

  • Severe liver impairment (Child-Pugh Class C, the most advanced grade of liver failure) or end-stage liver disease
  • Severe kidney impairment (creatinine clearance below 30 mL/min) or end-stage kidney disease requiring dialysis
  • Concurrent fluvoxamine therapy that cannot be substituted
  • Pregnancy and breastfeeding, on the basis of animal reproductive toxicity and absent human data
  • Known hypersensitivity to pirfenidone, including prior angioedema, or to any tablet or capsule ingredient

Risk Mitigation Strategies

  • Stepwise dose titration: Protocols start at 267 mg three times daily, rise to 534 mg in week two, then reach the full 801 mg. This blunts the first-month nausea and rash that drive most discontinuations.

  • Dosing with food, never fasting: Taking each dose mid-meal lowers peak concentration by roughly half and substantially reduces nausea and dizziness, at the cost of a modest absorption loss that does not offset the tolerability gain.

  • Scheduled liver monitoring: Alanine aminotransferase, aspartate aminotransferase and bilirubin are checked before starting, monthly for six months, then every three months, so that transient enzyme rises are caught and reversed before they force the cancer regimen to be interrupted.

  • Structured photoprotection: Daily broad-spectrum sunscreen at sun protection factor 50 or higher on all exposed skin, plus covering clothing and avoidance of midday sun and sunlamps, prevents the phototoxic burns that account for most skin-related withdrawals.

  • Vitamin D repletion alongside sun avoidance: Because photoprotection removes the main source of vitamin D, 25-hydroxyvitamin D is measured at baseline and every six months and supplemented to the 40–60 ng/mL range, preventing treatment-induced deficiency.

  • Weight and appetite tracking: Weight is recorded at every visit, with a 5% fall from baseline triggering dose reduction or nutritional support, since involuntary weight loss is the adverse effect most likely to derail cancer treatment.

  • Interaction audit before initiation: Every prescription, over-the-counter and supplement product is reviewed for CYP1A2 inhibitors and inducers and for photosensitisers, since these drive both the toxicity extremes and the silent loss of effect.

  • Temporary interruption rather than withdrawal: For moderate rash or enzyme rises, pausing one to two weeks and restarting a step down usually restores tolerance, avoiding permanent loss of a drug that cannot be re-titrated quickly.

Therapeutic Protocol

  • Standard antifibrotic dosing: The licensed antifibrotic dose is 801 mg three times daily, totalling 2403 mg per day, reached over a two-week titration. Oncology studies have used it at full strength or at reduced doses.

  • Reduced-dose radiation-injury regimen: The randomised radiation-injury trial used a lower schedule — 200 mg three times daily in week one, 300 mg in week two, then 400 mg three times daily through week 24 — alongside tapering steroids.

  • Perioperative schedule: In the lung cancer surgery protocol the drug began at 600 mg daily for two weeks, rose to 1200 mg daily, and surgery followed after at least two weeks at that dose.

  • Competing approaches: Two alternatives sit alongside it without a default. Nintedanib is the other licensed antifibrotic and blocks more tumour growth signals; corticosteroids alone remain the conventional treatment for radiation lung injury and for fibrosis flares.

  • Where each approach originated: The perioperative protocol came from the West Japan Oncology Group under Iwata and Yoshino; the combination programme from the Moffitt Cancer Center under Antonia; the radiation-injury regimen from Sun Yat-sen University Cancer Center.

  • Timing across the day: Doses are spread evenly across breakfast, lunch and dinner. Even spacing matters more than clock time because the short half-life produces deep troughs if doses cluster.

  • Half-life and dose splitting: The terminal half-life is roughly 2.4 hours, which is why three divided doses are mandatory. Single daily dosing produces a high peak, worse nausea, and no measurable drug for most of the day.

  • Genetic considerations: CYP1A2 activity dominates. No validated genotype-guided protocol exists, so smoking status is used as the practical proxy — smokers reach roughly half the expected exposure and are usually titrated to full dose faster.

  • Sex-based differences: Women reach higher exposure at the same dose through lower average CYP1A2 activity. No sex-specific dose is licensed, but slower titration and earlier dose reduction for nausea are commonly applied.

  • Age considerations: No age-based dose reduction is specified, and the registration population averaged near 70. Above 75, most practitioners extend each titration step to two weeks and monitor weight more closely.

  • Baseline biomarkers guiding the choice: Liver enzymes, kidney function and body weight determine whether full dosing is realistic. Baseline gas-transfer capacity determines whether the lung-protective rationale applies at all.

  • Pre-existing conditions: Moderate liver impairment calls for a reduced dose with fortnightly enzyme checks; inflammatory bowel disease calls for a slower titration; active radiotherapy to a skin-bearing field calls for deferral until the field settles.

Discontinuation & Cycling

  • Duration of use: In lung scarring the drug is taken indefinitely. In cancer settings it is deliberately time-limited — 24 weeks in the radiation-injury protocol, and from the perioperative window through surgical recovery in the surgery protocol.

  • Withdrawal effects: None are described. There is no dependence, rebound or discontinuation syndrome. What returns on stopping is the underlying fibrotic process, at the rate it would have progressed untreated.

  • Tapering: No taper is required pharmacologically, and abrupt cessation is standard when toxicity forces it. Where the drug is stopped electively, some practitioners step down through the titration levels to distinguish drug effects from disease effects.

  • Cycling: Cycling is not recommended and has never been tested. The mechanism depends on continuous suppression of a chronic signalling programme, and intermittent exposure would forfeit that while retaining the first-month tolerability penalty on each restart.

  • Restarting after interruption: After a pause exceeding 14 days, the full two-week titration is repeated from the lowest step, because gastrointestinal and skin tolerance is lost quickly and is not retained from the earlier course.

Sourcing and Quality

  • Prescription-only status: Pirfenidone is a prescription medicine everywhere it is licensed and is not a supplement. Any online seller offering it without a prescription is operating outside regulatory control, with no assurance of identity, dose or purity.

  • Branded and generic products: Esbriet is the originator brand in the United States and Europe. Generic pirfenidone tablets and capsules from established manufacturers became widely available after patent expiry and are therapeutically equivalent where approved by a stringent regulator.

  • Regional brands: Pirespa in Japan, Pirfenex in India, Etuary in China and Deskar in Mexico are established regional brands. Quality varies with the approving regulator rather than the brand name, so the marketing authorisation is the thing to check.

  • Formulation choice: Capsules of 267 mg and tablets of 267, 534 and 801 mg exist. Tablets reduce the daily unit count considerably at full dose, which matters when adherence competes with a cancer regimen already involving many tablets.

  • What to look for: A verifiable marketing authorisation from a stringent regulator, intact tamper-evident packaging, a batch number and expiry date, and dispensing through a licensed pharmacy. Compounded or repackaged pirfenidone offers no advantage and forfeits these guarantees.

  • Storage and handling: Labelled storage is below 30°C in the original light-resistant container. The molecule is light-sensitive, and extended decanting into clear organisers degrades it.

Practical Considerations

  • Time to effect: Lung-protective effects emerge slowly. The radiation-injury trial measured its endpoint at 24 weeks; the perioperative protocol required at least two weeks at full dose before surgery. Nothing meaningful happens within days.

  • Common pitfall — starting at full dose: Beginning at 801 mg three times daily without titration produces nausea severe enough that most people abandon the drug in the first fortnight, and tolerance cannot be recovered by simply persisting.

  • Common pitfall — dosing on an empty stomach: Taking doses between meals roughly doubles peak concentration and is the single most frequent avoidable cause of intolerable nausea and dizziness.

  • Common pitfall — unmanaged sun exposure: Photosensitivity is often discovered only after a severe burn. Protection belongs with the first dose rather than the first reaction.

  • Common pitfall — overlooking smoking: Continuing to smoke halves exposure, and stopping mid-course can raise it sharply. Neither change is visible without attention to symptoms, since drug levels are not routinely measured.

  • Regulatory status: Every oncology use described here is off-label. The licensed indication is idiopathic pulmonary fibrosis (progressive scarring of the lungs with no identified cause), and in some jurisdictions other progressive fibrotic lung disease.

  • Cost and accessibility: Generic entry cut the price sharply. Insurers and health systems have a structural incentive to favour a cheap generic over costly oncology agents, yet coverage stays tied to the licensed lung indication, which blocks the cheaper route.

Interaction with Foundational Habits

  • Sleep: Indirect and mildly disruptive. Insomnia and fatigue are both reported, and because CYP1A2 clears melatonin as well as pirfenidone, supplemental melatonin can linger and cause morning drowsiness. Moving the evening dose to dinner rather than bedtime, and halving any melatonin dose, usually resolves both.

  • Nutrition: Direct and clinically important. Every dose must be taken mid-meal to blunt nausea, which makes three substantial meals a practical requirement rather than a preference. Concentrated cruciferous extracts induce the clearing enzyme and lower drug levels; ordinary vegetable intake does not and carries no such penalty.

  • Exercise: Indirect. No effect on muscle protein synthesis or training adaptation is described, and the drug is not a performance concern. The practical constraints are dizziness in the first weeks, which argues against heavy loaded work immediately after a dose, and outdoor training, which requires the same photoprotection as any sun exposure.

  • Stress management: Indirect, with no measured effect on cortisol or the stress-response axis. The relevant interaction is behavioural: caffeine competes for the same clearing enzyme, so habitual high intake produces more tremor and palpitations on treatment, and halving intake at initiation is the usual adjustment.

Monitoring Protocol & Defining Success

Before the first dose, practitioners establish a liver panel, a full blood count, kidney function and — because the drug forces sun avoidance — vitamin D status. Where lung protection is the goal, baseline lung-function testing including gas transfer and forced vital capacity anchors everything that follows; where the drug is added to cancer treatment, baseline tumour imaging and body weight matter as much. Liver chemistry is then repeated monthly for the first six months and every three months thereafter, with lung function at three-month intervals and imaging on whatever schedule the cancer protocol already sets. Weight is checked at every visit, since appetite loss builds slowly and is easy to miss. Any dose change, or any newly added interacting drug, resets the monthly liver-testing clock.

Biomarker Optimal Functional Range Why Measure It? Context/Notes
Alanine aminotransferase 10–20 U/L (women), 10–25 U/L (men) Earliest signal of drug-related liver stress ALT = alanine aminotransferase. Conventional laboratories accept up to 40–55 U/L; the functional range is far tighter. Fasting not required. Action threshold is three times the upper laboratory limit
Aspartate aminotransferase 10–20 U/L (women), 10–25 U/L (men) Confirms and cross-checks a rise in the first enzyme AST = aspartate aminotransferase. Conventional laboratories accept up to 35–40 U/L; the functional range is far tighter. Also rises with muscle damage, so creatine kinase is the natural paired test after heavy exercise before a rise is attributed to the drug
Total bilirubin 0.3–1.0 mg/dL Distinguishes harmless enzyme rises from genuine liver dysfunction A rise here alongside enzymes is the combination that mandates stopping. Best drawn fasting; levels rise further after 12 hours without food in people carrying the common inherited bilirubin-processing variant
Estimated glomerular filtration rate Above 90 mL/min/1.73 m² The active metabolite is cleared by the kidneys eGFR = estimated glomerular filtration rate, a calculated measure of kidney filtering capacity. Conventional laboratories flag only values below 60 mL/min/1.73 m²; the functional target is higher. Below 30 mL/min the drug is avoided. Best paired with cystatin C where low muscle mass makes creatinine misleading
Body weight No established target; held within 5% of the individual’s own baseline Weight loss is the adverse effect most likely to interrupt cancer therapy Weighed at the same time of day, in the same clothing. A 5% fall triggers dose review regardless of absolute weight
Serum albumin 4.0–5.0 g/dL Marks the nutritional reserve available to absorb appetite suppression Conventional laboratories accept 3.5–5.0 g/dL; the functional floor is higher. Falls with inflammation as well as nutrition, so it is read alongside C-reactive protein (CRP, a general marker of inflammation) rather than alone
25-hydroxyvitamin D 40–60 ng/mL Enforced sun avoidance depletes it predictably Conventional laboratories call 30 ng/mL sufficient. Measured at baseline and every six months; supplements are absorbed best with the largest fat-containing meal
Gas-transfer capacity, percent predicted Above 80% predicted, or stable against the individual’s own baseline The endpoint on which the strongest human benefit was demonstrated DLCO = diffusing capacity of the lung for carbon monoxide. Correction for haemoglobin is essential, since it is frequently low during chemotherapy and will otherwise depress the result spuriously
Forced vital capacity, percent predicted Above 80% predicted, or stable against the individual’s own baseline Detects progression of underlying lung scarring FVC = forced vital capacity. Same-device, same-technician testing matters more than the absolute number; between-device variation exceeds the effect being tracked
Absolute neutrophil count Above 1.5 × 10⁹/L Guards the margin needed to continue chemotherapy safely Part of the full blood count. Pirfenidone does not suppress the bone marrow, but the concurrent cancer regimen usually does, and this is the shared decision point

Qualitative markers worth tracking alongside the laboratory values:

  • Appetite and the size of meals actually finished, rather than meals offered
  • Nausea severity in the hour after each dose, which reveals whether food timing is working
  • Skin reaction to ordinary daylight exposure, including through window glass
  • Breathlessness on a fixed, repeatable task such as one flight of stairs
  • Energy and daytime alertness, distinguishing drug fatigue from cancer-treatment fatigue by its timing relative to doses
  • Sleep onset and morning drowsiness, especially where caffeine or melatonin intake has not been adjusted

Emerging Research

For someone weighing an off-label repurposed drug against its tolerability cost, the trials below are the ones that will move the decision, because each replaces mechanistic reasoning with a human outcome.

  • Immunotherapy combination in lung cancer: NCT04467723 pairs pirfenidone with atezolizumab in second-line and beyond non-small cell lung cancer. Phase 1/2, 25 participants, University of Kansas Medical Center, with severe and life-threatening toxicity as the primary endpoints. Recruiting.

  • Colorectal cancer immune resensitisation: NCT06484153 adds pirfenidone to fruquintinib and a checkpoint antibody in advanced colorectal cancer whose DNA repair is intact — a setting where immunotherapy normally fails. Phase 2, 25 participants, progression-free survival primary. Recruiting.

  • Preventing radiation lung injury rather than treating it: NCT05704166 is the placebo-controlled prophylaxis counterpart to the treatment trial, in breast cancer patients receiving fewer, larger radiation doses. Phase 2, 214 participants, the largest placebo-controlled pirfenidone test in a cancer population under way. Recruiting.

  • Radiosensitisation in head and neck cancer: NCT06142318 tests the drug as a radiation sensitiser in head and neck squamous cell carcinoma, with objective response rate as primary endpoint. Phase 2, 66 participants, Nanfang Hospital. This is the first direct human test of the tumour-kill rationale.

  • Reversing established radiotherapy fibrosis: NCT06912763 is an adaptive platform trial at MD Anderson Cancer Center for radiotherapy-associated scarring of the mouth and jaw. Phase 2, 250 participants, safety primary. Recruiting.

  • The confirmatory surgical trial: The PIII-PEOPLE protocol of Sakairi et al., 2023 sets out the randomised phase 3 test of perioperative pirfenidone that the observational surgical data require. It carries a Japanese registry number rather than a ClinicalTrials.gov identifier.

  • What could weaken the case: The critique by Imai et al., 2025 argues the lung cancer incidence finding may be an artefact of index-date choice, detection bias and competing risks; the mouse liver model of Seniutkin et al., 2018 found antifibrotic effect without any reduction in tumour incidence, size or number.

  • What could strengthen the case: The preclinical combination work of Branco et al., 2025 reports antitumour activity with vinorelbine and carboplatin without added toxicity, supplying the dose and schedule rationale that the current lung cancer combination trials are built on.

Conclusion

Pirfenidone is an oral scar-suppressing medicine, originally used to slow lung scarring. Its relevance to cancer comes from the resemblance between scar-forming cells and the connective-tissue cells that tumours recruit to build their protective shell.

The strongest human signal is not an anti-tumour one. It is protective. In people whose lungs are already scarred or have been injured by radiation, the drug appears to restore the passage of oxygen from air into blood, cut the risk of sudden scarring flare-ups around lung surgery, and lower deaths in the months that follow. A large national records study also links its use to fewer new lung cancers, though a published critique argues that timing and detection effects could produce that pattern without real protection.

The anti-tumour claim itself remains laboratory work. Cell and animal studies show tumours becoming more sensitive to standard chemotherapy, immune cells regaining entry, and spread slowing; the only human tumours to shrink were nerve-sheath growths in a single small study, and no life has been lengthened. Against that sit dependable costs — nausea, rash, sun sensitivity, liver enzyme rises, and loss of appetite and weight — burdens that fall hardest on someone already thin and on treatment.

The evidence base is also lopsided. The safety and efficacy record was built and paid for by the company that owned the drug. Now that it is generic, no commercial sponsor stands to gain from funding large cancer trials, and no professional body earns from its use either way.

Top - Benefits - Risks - Protocol