POS0123 ASSOCIATION OF LUNG IMAGING PATTERN WITH PROGNOSIS AND IMMUNOSUPPRESSION RESPONSE IN CONNECTIVE TISSUE DISEASE ASSOCIATED INTERSTITIAL LUNG DISEASE
Bibliographic record
Abstract
Background Prognosis in connective tissue disease associated interstitial lung disease (CTD-ILD) is influenced by the underlying diagnosis and chest imaging pattern. Usual interstitial pneumonia (UIP), non-specific interstitial pneumonia (NSIP), and fibrotic hypersensitivity pneumonitis (fHP) patterns can be found across all CTD-ILD subtypes although their impact on disease evolution and treatment response is unclear. Objectives Our goal was to examine the association of lung imaging pattern with CTD-ILD progression, mortality, and immunosuppression response. Methods 615 patients with CTD-ILD enrolled in the Canadian Registry for Pulmonary Fibrosis had high-resolution chest computed tomography (HRCT) from their first ILD clinic visit reviewed in standardized multidisciplinary discussion. All CTD diagnoses were rheumatologist-confirmed. Experienced chest radiologists blinded to clinical data categorized each case into five groups: UIP, NSIP, organizing pneumonia (OP), fHP, and other patterns. Longitudinal percent-predicted forced vital capacity (FVC) and transplant-free survival were compared between imaging groups using linear mixed effects and Cox proportional hazards models adjusted for age, sex, smoking pack-years, and baseline FVC. Linear mixed effects models were used to compare pre- and post-treatment rate of FVC decline in patients with ≥6 months follow-up before and after treatment with mycophenolate, azathioprine, rituximab, cyclophosphamide, and/or tocilizumab. UIP was the reference group for all comparisons. Results The most frequent CTD subtypes were systemic sclerosis (SSc) (33%), rheumatoid arthritis (RA) (20%), and idiopathic inflammatory myopathy (IIM) (16%) with NSIP pattern present in 54% of all CTD-ILD (Table 1). On multivariable analyses among all CTD-ILD patients, NSIP was associated with a slower rate of FVC decline by 1.1%/year (0.2, 1.9) and a lower mortality HR (95%CI) of 0.65 (0.45, 0.93) compared to UIP. OP was also associated with a slower rate of FVC decline by 3.5%/year (2.0, 4.9) and a lower mortality HR (95%CI) of 0.18 (0.05, 0.57) compared to UIP. In contrast, fHP had a higher mortality HR (95%CI) of 1.58 (1.01, 2.40). The rate of FVC decline after treatment was not significantly different compared to pre-treatment in the UIP group but was slower in the NSIP group by 2.1%/year (1.4, 2.8). Subgroup analyses in RA-ILD and SSc-ILD showed the persistence of fHP having a higher mortality compared to UIP in RA-ILD. Conclusion The presence of an NSIP pattern was associated with improved outcomes and immunosuppression response compared to UIP in the overall CTD-ILD group. The findings of fHP associated with worse survival compared to UIP in CTD-ILD and in the RA-ILD are novel. These findings need to be further confirmed in disease specific cohorts and randomized trials of immunosuppression in patients with CTD-ILD. REFERENCES: NIL. Acknowledgements: NIL. Disclosure of Interests Boyang Zheng: None declared, Daniel-Costin Marinescu: None declared, cameron hague: None declared, Nestor Muller: None declared, darra murphy: None declared, Andrew Churg: None declared, Joanne Wright: None declared, Amna Al-Arnawoot: None declared, Ana-Maria Bilawich: None declared, patrick bourgouin: None declared, Gerald Cox: None declared, celine durand: None declared, Tracy Elliot: None declared, Jen Ellis: None declared, Jolene Fisher Consultant of: Boehringer-Ingelheim, AstraZeneca, Derek Fladeland: None declared, Amanda Grant-Orser: None declared, Gillian Goobie Grant/research support from: Boehringer Ingelheim, Zachary Guenther: None declared, Ehsan Haider: None declared, Nathan Hambly Speakers bureau: Boehringer Ingelheim, Grant/research support from: Boehringer Ingelheim, Janssen, Roche, James Huynh: None declared, Kerri Johannson Consultant of: Boehringer-Ingelheim, Hoffman-La Roche Ltd, geoff karjala: None declared, Nasreen Khalil: None declared, Martin Kolb Speakers bureau: Roche, Novartis, Boehringer Ingelheim, Grant/research support from: Boehringer Ingelheim, Pieris, Roche, Jonathon Leipsic Speakers bureau: GE Healthcare, Philips Healthcare, Stacey Lok Speakers bureau: Boehringer Ingelheim, sarah macisaac: None declared, micheal mcinnis: None declared, Helene Manganas Grant/research support from: Boehringer Ingelheim Canada, Hoffmann La Roche, Galapagos, BMS, Veronica Marcoux Grant/research support from: Astra Zeneca,Roche,Boehringer Ingelheim, John Mayo: None declared, julie morisset Speakers bureau: Roche, Boehringer Ingelheim, Ciaran Scallan: None declared, Tony Sedlic: None declared, shane shapera Consultant of: AstraZeneca, Boehringer Ingelheim, Hoffman LaRoche, Kelly Sun: None declared, victoria tan: None declared, Alyson Wong: None declared, Christopher Ryerson Speakers bureau: Boehringer Ingelheim, Hoffmann-La Roche, Astra Zeneca, Consultant of: Boehringer Ingelheim, Hoffmann-La Roche, Astra Zeneca.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.004 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".