e-Lung's Phase III Anchor Is Real, But Biomarker Qualification Gap Remains the Critical Unknown

Brainomix Unveils Phase III INBUILD® Quantitative CT Data

Brainomix announced new Phase III INBUILD® data demonstrating the potential of quantitative CT imaging to predict disease progression and assess treatment response in progressive pulmonary fibrosis. This landmark study, conducted in collaboration with Boehringer Ingelheim, highlights the utility of AI-powered imaging solutions, such as Brainomix's e-Lung, in clinical trials and patient management for interstitial lung diseases (ILD). The findings suggest that quantitative CT can serve as a valuable biomarker endpoint, potentially improving the assessment of therapeutic efficacy and patient outcomes in this challenging condition. The data is expected to be published in a reputable journal like The Lancet Digital Health.

• The new Phase III INBUILD® data specifically focuses on the application of quantitative CT imaging, demonstrating its significant potential in predicting the trajectory of disease progression and accurately assessing how patients with progressive pulmonary fibrosis respond to treatment. This represents a crucial advancement in leveraging AI-driven diagnostics for more precise patient stratification and monitoring in interstitial lung diseases.
• The findings from the INBUILD study underscore the value of AI-powered imaging solutions, such as Brainomix's e-Lung, as robust biomarker endpoints in clinical trials. By providing objective and quantifiable measures of lung changes, these tools can enhance the efficiency and accuracy of drug development for conditions like progressive pulmonary fibrosis, potentially accelerating the availability of effective therapies and improving patient care.
• The landmark INBUILD study, which generated this significant data, was conducted in collaboration with Boehringer Ingelheim, highlighting a successful partnership in advancing medical imaging and therapeutic assessment. The data's potential publication in a prestigious journal like The Lancet Digital Health further validates its scientific rigor and clinical importance, ensuring broad dissemination within the medical community.

The Unmet Need for Precise PPF Progression Assessment

Despite the availability of antifibrotic therapies, the management of Progressive Pulmonary Fibrosis (PPF) remains constrained by significant therapeutic gaps — from the absence of disease-reversing agents to the unpredictability of disease course across subtypes. Current treatment strategies slow progression but fall short of addressing the full complexity of the condition.

• Antifibrotics reduce but do not reverse functional decline. Pirfenidone and nintedanib have demonstrated reductions in lung function decline, but neither reverses existing lung damage nor fully addresses the complex pathophysiology of pulmonary fibrosis. Treatment options for IPF, the most debilitating ILD subtype, have remained mostly unchanged in the last decade.

• Therapeutic response varies significantly by disease subtype and radiological pattern. In a multicenter retrospective cohort of 222 PPF patients, the progression rate was significantly higher in patients with usual interstitial pneumonia (UIP) (P = 0.003), and the effect of antifibrotic drugs on progression varied especially among radiological patterns. CTD-ILD had the longest survival time (166.5 ± 9.2 months) while fibrotic hypersensitivity pneumonitis had the shortest (87.6 ± 9.2 months) (P = 0.011), underscoring the need for an individualized approach.

• No proven therapies exist for acute exacerbations of ILD. Corticosteroids remain a mainstay despite the absence of clear evidence of benefit. Immunosuppressant agents — cyclophosphamide, cyclosporine A, and tacrolimus — have purported success only in historical cohort studies. The sole published randomized controlled trial, studying recombinant thrombomodulin for IPF exacerbation, did not meet its primary outcome of survivor proportion at 90 days.

• Tolerability of antifibrotic agents poses a real-world adherence challenge. Diarrhea is the most frequent adverse event with nintedanib; among patients in INBUILD-ON who experienced diarrhea, 90.0% had only mild or moderate events, yet adverse events led to discontinuation at a rate of 16.7 per 100 patient-years. In elderly CTD-ILD patients, gastrointestinal adverse events occurred in 39.0% of those receiving antifibrotics (26.8% diarrhea, 22.0% nausea), with abnormal liver function observed in 17.1%.

• Reliable biomarkers for predicting PPF progression remain an unmet clinical need. In a retrospective cohort of 203 non-IPF ILD patients, only 23.6% developed PPF during follow-up, yet identifying which patients will progress prospectively remains difficult. While serum CYFRA21-1 (optimal cut-off 5.12 ng/mL, AUC = 0.712) and proteomic signatures have shown promise, these require validation in prospective multicentre studies before entering routine clinical practice.

• Evidence base for non-IPF and CTD-ILD subtypes remains limited. Outside the Scleroderma Lung Studies (SLS I and II) confirming effectiveness of cyclophosphamide and mycophenolate mofetil in SSc-ILD, there is little data on the treatment of other CTD-ILD subtypes. Prospective, randomized clinical trials are urgently needed to assess the real impact of antifibrotic therapy across the heterogeneous spectrum of progressive fibrosing non-IPF ILDs.

Integrating Quantitative CT into the Landmark INBUILD Study

The INBUILD trial evaluated nintedanib in patients with progressive pulmonary fibrosis (PPF), demonstrating that nintedanib slowed the decline in forced vital capacity (FVC) versus placebo, with a safety profile characterised mainly by gastrointestinal events. INBUILD-ON, the open-label extension (ClinicalTrials.gov NCT03820726), assessed the safety of nintedanib during longer-term treatment and collected data on FVC.

Quantitative CT: An Emerging Endpoint for Progressive Pulmonary Fibrosis

Clinical trials and real-world studies in PPF have moved beyond traditional lung function metrics to incorporate a broader set of endpoints that capture disease burden across functional, radiographic, patient-reported, and survival dimensions.

• Composite disease progression endpoints: The phase 2 admilparant study defined disease progression as a composite of relative decline of ≥10% in percentage of predicted forced vital capacity (ppFVC), acute exacerbation, all-cause hospitalization, and all-cause mortality, assessed via a Kaplan-Meier product-limit approach over 26 weeks. In both the IPF and PPF cohorts, the most frequent first event within this composite was a relative decline of ≥10% in ppFVC.

• Time-to-event endpoints encompassing acute exacerbation, hospitalization, and death: The FIBRONEER-ILD trial designated time to first acute exacerbation of ILD, hospitalisation for respiratory cause, or death as a key secondary endpoint. For nerandomilast, hazard ratios versus placebo were 0.78 (95% CI 0.61, 1.00) for 9 mg bid and 0.77 (95% CI 0.60, 0.99) for 18 mg bid; for death specifically, the hazard ratio was 0.51 (95% CI 0.34, 0.78) for both doses.

• Radiographic progression as a primary endpoint: A two-center Hungarian real-world study used blinded assessment of paired high-resolution computed tomography (HRCT) scans as its primary endpoint — the proportion of patients without significant radiographic progression at 6–12 months — finding radiographic stability in 65.8–78.9% of patients, with improvement most commonly seen in ground-glass opacities and traction bronchiectasis remaining largely unchanged.

• Patient-reported outcomes and workplace productivity: The ILD-PRO Registry assessed workplace productivity loss via the Work Productivity and Activity Impairment (WPAI) questionnaire, finding a mean (SD) percentage of working hours lost of 24.4 (30.2)% and an estimated mean (SD) annual cost of productivity loss of $24,815 ($17,917) among patients with any productivity loss. Lower FVC % predicted, higher St. George's Respiratory Questionnaire scores, and lower Cough and Sputum Assessment Questionnaire scores were associated with greater odds of productivity loss.

• Cough-specific quality of life: The adaptive Bayesian TCM trial in PPF designated change in Leicester Cough Questionnaire (LCQ) score as its primary patient-reported outcome, reflecting cough-specific quality of life. A separate 6-month feasibility study further explored objective cough monitoring via wearable devices, correlating cough intensity with LCQ and King's Brief Interstitial Lung Disease (K-BILD) questionnaire scores.

• Home-based digital monitoring endpoints: The eurILDreg ASA-ILD algorithm integrates home spirometry, pulse oximetry at rest and during exercise (one-minute sit-to-stand test), and patient-reported outcome measures (EQ5D VAS, K-BILD, LCQ) via a smartphone application, positioning digitally captured functional and symptomatic data as emerging endpoints for early detection of ILD progression and acute exacerbations.

Quantitative CT: A New Horizon for Progressive Pulmonary Fibrosis Assessment

The recent announcement regarding new Phase III INBUILD® data, highlighting the potential of quantitative CT (qCT) imaging in progressive pulmonary fibrosis (PPF), signals a transformative moment for interstitial lung disease (ILD) management and drug development. While forced vital capacity (FVC) has long been the primary measure, the INBUILD data, leveraging AI-powered solutions like Brainomix's e-Lung, suggests qCT can serve as a valuable, objective biomarker endpoint, offering a more nuanced understanding of disease trajectory.

This shift is not merely incremental; it represents a fundamental re-evaluation of how we measure disease. Research indicates that qCT, particularly when combined with FVC changes, significantly improves the stratification of long-term outcomes in IPF. It can quantify fibrosis extent and vascular involvement, distinguishing progression in treated versus untreated patients, which is crucial for understanding therapeutic impact. The integration of qCT radiomics with clinical features, even through machine learning, has demonstrated robust predictive performance for fibrosis progression, pointing towards a future of more precise risk assessment and intervention.

For pharmaceutical companies, this development carries significant strategic implications. The ability to utilize qCT as a primary or co-primary endpoint in clinical trials could:

• Streamline drug development: Potentially accelerating trials by providing earlier and more sensitive signals of efficacy.

• Refine patient stratification: Enabling more targeted recruitment of patients likely to benefit from specific therapies, enhancing trial success rates and informing personalized medicine.

• Differentiate market offerings: Companies integrating validated qCT solutions into their development programs could gain a competitive edge.

However, the path to widespread adoption is not without its challenges. The literature points to significant hurdles, including the lack of standardized image acquisition, reconstruction parameters, and reporting methods across different centers. This heterogeneity can impede direct comparisons and meta-analyses, posing risks to regulatory approval and broad clinical utility. Furthermore, much of the existing evidence for qCT's prognostic value comes from retrospective studies, underscoring the need for robust prospective validation. Addressing governance concerns and ensuring appropriate algorithm anchoring for AI-powered qCT tools will be paramount to building trust and facilitating their integration into routine clinical practice. International collaboration is already underway to tackle these issues, paving the way for qCT to fulfill its promise in revolutionizing ILD care.