Disease indication · a manifestation of rheumatoid arthritis
The same patient, the same autoimmunity, the same seven drugs — and in this organ almost none of them work. The one therapy that helps came from a different disease entirely.
Disease-modifying: Slow the loss of lung function, judged on forced vital capacity — an objective, spirometric measure that cannot be moved by how a patient reports feeling. One therapy has cleared that bar, and it is an antifibrotic borrowed from idiopathic pulmonary fibrosis rather than any of the immune mechanisms that solved the joints.
Nintedanib is approved for chronic fibrosing interstitial lung disease with a progressive phenotype — a category this disease qualifies for. No therapy has ever been approved for RA-associated ILD as such.
None of the seven mechanisms approved for rheumatoid arthritis carries a lung indication. TNF blockade, the most transformative of them in the joints, is generally avoided in patients with progressive lung disease.
A leading cause of death in rheumatoid arthritis, behind cardiovascular disease. Median survival from diagnosis is commonly reported between three and eight years depending on cohort and radiographic pattern, with UIP-pattern disease at the worse end.
Annual rate of FVC decline over 52 weeks. Trials frequently enrol across several fibrotic ILD causes at once rather than in this disease alone, because recruiting pure RA-ILD cohorts at scale is impractical.
This page covers the lung half of rheumatoid arthritis. The joint half is a different story entirely — seven approved mechanisms and structural damage that can be halted outright — and holding the two side by side is the point: same patient, same autoimmunity, same drugs, opposite outcomes.
Inflammatory synovitis eroding cartilage and bone — the disease as it is usually described, and the half that modern treatment transformed.
Progressive interstitial fibrosis, affecting roughly one patient in ten clinically and far more on imaging. Everything else on this page is about this half.
You are here
The two halves may not be one disease in the way the shared name implies. The strongest known genetic risk factor for the lung manifestation is the MUC5B promoter variant — the dominant common-variant risk factor for idiopathic pulmonary fibrosis, with no role in joint disease.
This is the diagnostic case for the rest of the site. Parkinson's is stuck partly because it has no accepted endpoint; here there is one — FVC decline, the same measure that carried two antifibrotics to approval in idiopathic pulmonary fibrosis. A good endpoint is necessary and clearly not sufficient.
| Endpoint | Type | Read at | Standing |
|---|---|---|---|
| Forced vital capacity (FVC) decline | Performance | 52 weeks | Accepted |
| Progressive fibrosing phenotype (composite) | Composite | 24–52 weeks | Accepted |
| Diffusing capacity for carbon monoxide (DLCO) | Performance | 52 weeks | Contested |
| Quantitative HRCT fibrosis extent | Imaging | 52 weeks | Exploratory |
| Six-minute walk distance | Performance | 8–52 weeks | Exploratory |
Compare this matrix against the joint page's. There, models with poor construct validity predicted the clinic anyway. Here the standard model is a single acute chemical lung injury that resolves on its own — and its predictive record is the worst on this site for a disease that nonetheless has an approved drug.
| Model | FaceDoes it look like the disease? | ConstructDoes it arise the same way? | PredictiveSlow lung decline |
|---|---|---|---|
The striking column is not the failures but the mechanisms: of the seven that transformed the joints, none carries a lung indication, one is actively avoided, and the only approved therapy came from idiopathic pulmonary fibrosis.
| Intervention | Rested on | Primary endpoint | Reached | Outcome |
|---|---|---|---|---|
Abatacept Bristol Myers Squibb / investigator-led · CTLA4-Ig blockade of T-cell costimulation | SKG (lung) | Change in FVC percent-predicted; prospective studies in this population remain limited | Phase 2 / observational 2024 | Ongoing |
Pirfenidone Genentech / investigator-led · Antifibrotic with broad effects on TGF-β signalling and fibroblast collagen synthesis | BleomycinAdTGF-β1 | Composite of decline in FVC percent-predicted of at least 10%, or death, at 52 weeks | Phase 2 TRAIL1 · 2023 | Missed primary |
Rituximab Investigator-led · Anti-CD20 B-cell depletion | Change in FVC at 24 weeks, against intravenous cyclophosphamide as active comparator | Phase 2b RECITAL · 2023 | Benefit, no approval | |
Cyclophosphamide Investigator-led · Broad cytotoxic alkylating immunosuppression | Change in FVC at 24 weeks, as the active comparator arm in RECITAL | Phase 2b comparator RECITAL · 2023 | Benefit, no approval | |
Tocilizumab Genentech / Roche · Interleukin-6 receptor blockade | Change in FVC percent-predicted at 48 weeks — in systemic sclerosis, not in this disease | Approved in a different disease focuSSced · 2021 | Ongoing | |
Nintedanib Boehringer Ingelheim · Intracellular inhibition of multiple tyrosine kinases, including PDGF, FGF and VEGF receptors, reducing fibroblast proliferation | Bleomycin | Annual rate of decline in FVC, in millilitres per year, over 52 weeks | Approved INBUILD · 2020 | Approved |
Mycophenolate mofetil Investigator-led · Inhibition of lymphocyte proliferation via IMPDH blockade | No randomised controlled trial with a lung endpoint has been completed in this disease specifically | Observational only 2013 | Benefit, no approval | |
TNF inhibitors Multiple · Neutralisation of tumour necrosis factor | SKG (lung) | No prospective controlled trial has been run with a lung endpoint; evidence is observational, from registries and cohorts | Observational only 2010 | Failed |
The total volume of air a patient can forcibly exhale after a maximal breath in, tracked over time
The most important fact on this page: unlike Parkinson's, this disease has a fully accepted, objective, structural endpoint — and it is still largely unsolved. A good endpoint is necessary and not sufficient.
Cheaper per programme than the joint disease — the trials are smaller, because recruiting patients with both rheumatoid arthritis and progressive lung fibrosis is hard — and more expensive per success, because far fewer of these programmes work.
Fifty-two weeks of serial spirometry to fit a rate of decline. Smaller than the equivalent joint-disease programme, because recruiting patients with both rheumatoid arthritis and progressive lung fibrosis is genuinely hard — trials routinely enrol across several fibrotic ILD causes at once to fill.
788 patient-years against 250 for a symptomatic programme. Cost grows more slowly than exposure because the fixed enrolment cost of each participant — screening, baseline imaging — is paid once however long they are then followed. The goal, not the molecule, still sets the budget.
Citedenrolment, duration and study counts, from named trials.Assumptionthe two cost parameters below.Computedeverything else.
| Phase | Design | N × studies | Duration | Patient-years | Modelled cost |
|---|---|---|---|---|---|
| Phase 1 | SAD / MAD, healthy volunteers Reads: Safety, tolerability and pharmacokinetics Antifibrotics carry gastrointestinal and hepatic tolerability questions that shape later dosing, so first-in-human work is not abbreviated. Conventional first-in-human designmoderate confidence | 50 | 9 mo | 38 | $1.4M–$2.2M |
| Phase 2 | Dose-ranging on rate of FVC decline Reads: Annual rate of FVC decline, with HRCT fibrosis extent as a supporting read A full year even at Phase 2, because the endpoint is a rate of change rather than a level — there is no shorter read available, unlike the twelve-week ACR response that serves the joint disease. Antifibrotic Phase 2 programme designs in fibrotic ILDmoderate confidence | 150 | 1 yr | 150 | $4.8M–$8.0M |
| Phase 3 | Two pivotals, 52 weeks, serial spirometry Reads: Annual rate of decline in FVC in millilitres per year, over 52 weeks Sized after INBUILD and TRAIL1. Roughly a third the enrolment of a joint-disease pivotal — not because the endpoint is easier, but because the eligible population is a fraction the size and progressive cases are harder still to find. Flaherty et al., NEJM 2019 (INBUILD); Solomon et al., Lancet Respir Med 2023 (TRAIL1)moderate confidence | 300 × 2 | 1 yr | 600 | $19M–$32M |
| Total | 2.8 yr of clinical development, treating phases as sequential | 788 | $25M–$42M |
Model assumptions, not citations — the only free parameters here. The fixed component covers screening, baseline imaging and randomisation, incurred once per participant; the running component covers visits, monitoring and site fees for each year on study. Splitting them matters: a single rate per patient-year would price short symptomatic trials far below what pivotal trials actually cost.
Calibration: priced this way a single pivotal trial in this programme comes to $4.4M–$7.1M, against published pivotal trial costs of $12–33M (median $19M) across all indications. A supportive-care trial here is small, short and often unblinded by necessity — you cannot placebo-control a rehabilitation programme — so it lands far below the published range. That range is set by the 52-week antifibrotic programmes in the disease-modifying archetype. Moore et al., JAMA Internal Medicine 2018.
Programme cost divided by a 9% likelihood of approval from Phase 1 in fibrotic interstitial lung disease — what one success costs once the failures are paid for. BIO / Informa / QLS, Clinical Development Success Rates — respiratory sits mid-table overall, and fibrotic lung disease sits well below it.
Read this one carefully: The lowest rate on this site outside Parkinson's, and for a reason this page's model matrix makes explicit: the standard preclinical model produces fibrosis that resolves on its own, so it has returned positive results for a long list of compounds that then failed in patients. Two antifibrotics have cleared this bar in twenty years. Read 9% as reflecting the model problem, not a lack of effort or money.
These are portfolio figures spanning all of R&D and already carry the cost of failure, so they are not comparable with the per-programme clinical costs above — they are the order of magnitude those costs roll up into.
Two shifts: treating fibrosis as its own disease regardless of what triggered it, and a genuine reversal of what the field believed about its most-used drug.
The progressive fibrosing phenotype construct treats scarring lungs as one entity regardless of what started them scarring, whether autoimmunity, environmental exposure or nothing identifiable. That reframing is the only reason this disease has an approved therapy — it inherited one from idiopathic pulmonary fibrosis rather than generating its own. It also implies the reverse: this disease may be better served by respiratory drug development than by rheumatology.
INBUILD (Flaherty et al., NEJM 2019); FDA approval of nintedanib for progressive fibrosing ILD, 2020
Methotrexate, the anchor drug of rheumatoid arthritis, was long suspected of causing or worsening lung disease — a belief that led clinicians to withhold it from exactly the patients most at risk. More recent cohort work points the other way, associating it with delayed onset of interstitial disease, while keeping acute methotrexate pneumonitis as a distinct and genuinely drug-induced entity. A rare, well-documented case of a field inverting a long-held safety assumption.
Juge et al., European Respiratory Journal 2021; Kiely et al., BMJ Open 2019
The strongest genetic risk factor for this manifestation is the MUC5B promoter variant, which is also the dominant common-variant risk factor for idiopathic pulmonary fibrosis and has no role in joint disease. If the lung manifestation is genetically closer to a fibrotic lung disease than to the arthritis it accompanies, then expecting the drugs that fixed the joints to fix the lungs was misplaced from the start.
Juge et al., NEJM 2018 (MUC5B variant and RA-ILD)