Clean Sweep or Just a Wash? What BAL Lymphocytes Really Say About Steroid Outcomes in ILD

While research continues to reshape our understanding of interstitial lung disease therapies, we still lack the tools to predict who will benefit from treatments, including immunosuppression. In this regard, idiopathic pulmonary fibrosis (IPF) continues to remain the outlier because treatment of IPF with immunosuppressive therapies has been associated with more harm than good [1]. For those with non-IPF ILDs, exacerbations are often treated with corticosteroids. However, corticosteroids are not always the right answer because of the risk of adverse effects and the uncertainty of the outcomes. Bedside tools that separate the steroid-sensitive ILDs from non-steroid-responsive ILDs can turn guesswork into more guided decision-making. Leveraging bronchoalveolar lavage (BAL) samples to delineate ILD subtypes remains a focal point of research, though the endeavour continues to pose significant challenges due to disease heterogeneity and diagnostic complexity. In particular, BAL lymphocytes percentage (BAL LP) has also been studied in three different scenarios: 1. diagnosing ILD subtypes; 2. predicting prognosis and survival; and 3. response to immunosuppressive treatment with variable results [2-4]. Furthermore, BAL is rarely performed in the United States for the diagnosis of ILD, which limits the potential generalisability of BAL research in ILD. The recent prospective study published by Dhooria et al. [5] in Respirology looked at the utility of BAL lymphocyte percentage to shed a beacon of light on this clinical conundrum. They included a group of 150 non-IPF ILD patients to understand the relationship between BAL lymphocyte percentage (BAL LP) and early glucocorticoid response (EGR). EGR was measured by improvement in symptoms and pulmonary function within 2 months of starting steroids. The study found that a higher BAL LP was associated with EGR. Patients with higher BAL LP were more likely to show EGR after steroid initiation. The team found that the response rate was 49% in those with a BAL lymphocyte percentage 10%–19%. The response rate improved to 75% in those with a BAL LP > 40%. They also found that BAL LP < 20% but a history of significant smoke exposure (cigarettes, bidis, or cooking biomass fuel) predicted disease progression risk even independent of steroid use. Not surprisingly, the authors also discovered that a shorter duration of respiratory symptoms was independently tied to favourable steroid response. Symptoms less than 6 months plus a BAL LP > 20% had a more robust EGR based on improvement in FVC. This finding reinforces the time-sensitive nature of anti-inflammatory therapy in ILD. While inflammatory CT patterns were more common among responders, nearly half (46%) of these patients did not respond, reiterating that while radiological phenotypes are important, they are not independently diagnostic. With careful execution, Dhooria et al. have produced a study that both clarifies and complicates our understanding—inviting many fresh questions for the community. The question now arises if we should standardise the use of BAL samples in patients with ILDs to help categorise responsiveness to steroids. Based on this study's results, BAL could be a good starting point for non-IPF ILD. If a patient has a higher BAL LP > 20%, steroids may be more beneficial and should be initiated. Conversely, this also could be helpful in patients with lower BAL LP to minimise the use of steroids. Additionally, the fundamental principles of medicine, a complete and thorough history, particularly asking about smoke exposure, can give us a better idea about what their disease phenotype may look like. The data also reaffirm that radiological findings do not consistently predict steroid responsiveness. Thus, while it may be resource intensive, we can turn our focus to bronchoscopy to obtain BAL samples, if available, rather than CT findings, to provide helpful information and an actionable treatment plan. As with any single-centre study, broader validation in diverse and multicentre cohorts is needed. While inclusion of various ILD subtypes in this study increases its generalizability, we must also remember that this creates more room for complexity and confounding. Some patients may also be too sick or have too large of an oxygen requirement to safely get a bronchoscopy. We cannot rely on the idea of a 'one size fits all' when utilising steroids in patients with diverse phenotypes of ILD. Important examples in the literature include greater typing of lymphocytes including B or T cell dominant subtypes, as well as protein biomarkers from the BAL [6]. Ultimately, BAL to assess lymphocyte measurement with clinical reasoning could offer a pragmatic step towards more personalised precision medicine. The quest for new discovery for diagnosis and treatment of ILD continues, and the era of empiric steroids for all-comers may be replaced by a more personalised approach to stratify patients by BAL lymphocyte percentage. This may help clinicians determine if steroid treatment is an appropriate treatment in patients who present with ILD. As it turns out, with more data, the secret to steroid responses may have been in our grasp—literally at our bronchoscope's tip. D.J.K. is an Editorial Board member of Respirology and a co-author of this article. He was excluded from all editorial decision-making related to the acceptance of this article for publication. J.V. has no conflicts of interest to declare.