SMARCA4-deficient non-small cell lung cancer and immunotherapy resistance: mechanisms and novel strategies-a mini-review.

Introduction
SMARCA4-deficient thoracic tumors represent a distinct and aggressive subset of malignancies characterized by inactivating mutations in the SMARCA4 gene, which encodes the BRG1 protein, a critical ATPase subunit of the SWI/SNF chromatin remodeling complex [1, 2]. The most prevalent and clinically significant manifestation is SMARCA4-deficient non-small cell lung cancer (SMARCA4-dNSCLC). The same defining alteration also drives the highly aggressive SMARCA4-deficient undifferentiated thoracic tumor (SMARCA4-UT), a rarer entity recognized as distinct in the latest WHO classification [3, 4]. This review focuses on SMARCA4-dNSCLC but, where relevant, draws upon evidence from SMARCA4-UT to illuminate the core biological consequences—such as epigenetic dysregulation and immune evasion—that are fundamental to the therapeutic challenge across SMARCA4-deficient cancers. SMARCA4 alterations occur in approximately 8–10% of NSCLCs, with a higher prevalence in certain subtypes such as mucinous lung adenocarcinomas [2, 5, 6].
The SWItch/Sucrose non fermentable (SWI/SNF) complex plays a fundamental role in regulating chromatin structure and gene expression by utilizing ATP hydrolysis to provide energy for chromatin remodeling [1, 2]. Deficiency in SMARCA4 disrupts this process, leading to widespread epigenetic dysregulation that contributes to tumorigenesis, disease progression, and therapeutic resistance [7–9]. Clinically, SMARCA4-dNSCLC is an aggressive disease with limited therapeutic options, a point elaborated in the following sections. Although immune checkpoint inhibitors (ICIs) have revolutionized advanced NSCLC treatment, patients with SMARCA4-deficient tumors often respond poorly [7, 9, 10]. This resistance is multifactorial, involving both tumor-intrinsic factors and an immunosuppressive microenvironment [9, 11, 12]. Understanding the mechanisms underlying immune evasion in these tumors is crucial for developing effective therapeutic strategies.
This review comprehensively examines the current state of knowledge regarding SMARCA4-deficient thoracic tumors, with a particular focus on the molecular characteristics that influence response to immunotherapy, the tumor microenvironment features that contribute to immune evasion, and the clinical efficacy of immune checkpoint inhibitors. Furthermore, we explore emerging therapeutic strategies beyond conventional immunotherapy that hold promise for overcoming treatment resistance in this challenging disease subset.

Clinicopathological and molecular characteristics of SMARCA4-deficient NSCLC
SMARCA4-deficient thoracic tumors exhibit distinct clinicopathological features. They predominantly affect older males with a heavy smoking history (approximately 90% of cases) and typically present at advanced stages with frequent metastases [13–15]. A detailed analysis of metastatic patterns reveals that among patients with SMARCA4-dNSCLC, approximately 66% present with distant metastases at diagnosis. The most common sites include bone (30%), pleura (26%), intrapulmonary sites (23%), brain (16%), and adrenal glands (15%). Notably, the frequency of adrenal metastasis is significantly higher in SMARCA4-dNSCLC compared to SMARCA4-intact NSCLC, underscoring a specific propensity for subdiaphragmatic spread in this subtype [14]. The median age at diagnosis ranges from 60 to 70 years, though SMARCA4-UTs may occur in slightly younger patients [3, 4].
Pathologically, SMARCA4-dNSCLC is characterized by solid growth patterns and poorly differentiated morphology [3, 4, 15]. This stands in contrast to the related SMARCA4-UT, which exhibits fully undifferentiated or rhabdoid features, underscoring the spectrum of differentiation states linked to SMARCA4 loss. Immunohistochemically, loss of BRG1 protein expression is a hallmark of both [13, 15, 16]. SMARCA4-dNSCLC is frequently negative for TTF-1 and p40 but positive for CK7 [13, 14], and typically retains some epithelial markers. The frequent expression of neuroendocrine (Syn, CD56) and stem cell markers (SALL4, SOX2) in SMARCA4-UT [3, 17] further highlights the divergent phenotypic programs that can emerge from the common genetic background, with SMARCA4-dNSCLC representing a more epithelial-anchored malignancy. PD-L1 expression demonstrates heterogeneous distribution, with approximately 33–55% of cases showing low expression and 11% exhibiting high expression [13, 18].
Molecular profiling reveals that SMARCA4-deficient tumors harbor a distinct genomic landscape. TP53 mutations are almost universally present, occurring in 90–100% of cases. Other frequent co-mutations include LRP1B (67%), KEAP1 (50%), STK11 (33%), and KRAS (17%) [3, 13, 19]. The pattern of SMARCA4 mutations themselves has clinical significance, with truncating mutations (class I) associated with complete loss of BRG1 protein expression and worse prognosis compared to missense mutations (class II) [6, 19]. SMARCA4-deficient tumors typically lack targetable drivers in EGFR, ALK, and ROS1, limiting options for molecularly targeted therapies [15, 16, 19]. The complex interplay between these genomic alterations creates a challenging therapeutic landscape that necessitates a nuanced approach to treatment selection.
Beyond its established role in immunotherapy resistance, emerging evidence suggests that SMARCA4 deficiency may also modulate responses to molecularly targeted therapies. For instance, in KRAS G12C-mutant lung cancers, concurrent SMARCA4 alterations have been associated with inferior outcomes following treatment with KRAS G12C inhibitors, underscoring a broader role in therapeutic resistance beyond the immune context [20]. Furthermore, while SMARCA4-deficient tumors typically lack classical targetable drivers such as EGFR, ALK, or ROS1, the impact of SMARCA4 loss on the efficacy of targeted agents against other rare or emerging actionable alterations (such as RET fusions, MET amplifications) remains an important area for future investigation. This highlights SMARCA4 deficiency as a potential cross-therapeutic resistance modifier, influencing outcomes across both immunotherapeutic and targeted treatment paradigms.

The tumor microenvironment and immune evasion mechanisms in SMARCA4-deficient tumors
The tumor microenvironment (TME) of SMARCA4-dNSCLC exhibits distinct characteristics that contribute significantly to its aggressive behavior and resistance to therapy. Studies across the spectrum of SMARCA4-deficient tumors, including the extreme phenotype of SMARCA4-UT, have consistently revealed that these tumors frequently display an “immune-desert” phenotype, characterized by minimal T-cell infiltration and poor immune activation [11, 12]. This immunosuppressive microenvironment represents a critical barrier to effective immunotherapy and understanding its underlying mechanisms is essential for developing strategies to overcome treatment resistance.
Several studies have demonstrated that SMARCA4-deficient tumors are highly infiltrated by immunosuppressive cell populations, particularly FOXP3 + regulatory T cells and neutrophils [11]. This abundance of regulatory immune cells creates an environment that suppresses effective anti-tumor immunity and promotes tumor progression. The density of CD8 + cytotoxic T cells, in contrast, is not significantly different from that in SMARCA4-intact tumors, suggesting that the quality rather than the quantity of T-cell infiltration may be impaired [11]. Additionally, tertiary lymphoid structures (TLS), which are associated with improved responses to immunotherapy in other cancer types, are notably absent in most SMARCA4-deficient tumors [12].
Molecularly, SMARCA4 deficiency profoundly reprograms the enhancer landscape, downregulating key components of the innate immune system [7]. Specifically, SMARCA4 loss causes marked reduction in chromatin accessibility at enhancers associated with STING1, IL1β, type I interferon, and inflammatory cytokines. This epigenetic reprogramming impairs the recruitment and activation of dendritic cells and CD4 + T cells, further contributing to the immunosuppressive TME. The transcription factor NF-κB appears to play a crucial role in this process, as SMARCA4 and NF-κB co-occupy genomic loci associated with immune stimulation.
The co-occurrence of specific mutations further shapes the immune landscape of these tumors. STK11 and KEAP1 mutations, which are frequently present in SMARCA4-deficient tumors, are known to drive immune evasion through multiple mechanisms [9, 21]. STK11 mutations promote an immune-excluded TME with reduced T-cell infiltration, while KEAP1 mutations, through constitutive NRF2 activation, facilitate oxidative stress adaptation and metabolic reprogramming that supports immune evasion [9]. The combination of these mutations creates a particularly immunosuppressive environment that is highly resistant to ICIs [9, 10, 22].
Antigen presentation mechanisms may also be compromised in SMARCA4-deficient tumors. Transcriptomic analyses have revealed downregulation of genes involved in antigen processing and presentation, including components of the MHC class Ib pathway such as CD1A, CD1C, and CD1E [23]. This impaired antigen presentation capacity further limits the ability of the immune system to recognize and target these tumors effectively.

Predictive biomarkers and genomic correlates of response to immunotherapy
Identifying reliable predictive biomarkers for immunotherapy response in SMARCA4-deficient thoracic tumors remains challenging due to the complex interplay of genomic and microenvironmental factors. To systematically evaluate the existing and emerging evidence, this section is organized around three broad categories of biomarkers: (a) conventional biomarkers, (b) genomic co-alterations, and (c) emerging or exploratory biomarker approaches.

Conventional biomarkers: PD-L1 and TMB
Current evidence suggests that conventional biomarkers such as PD-L1 expression and tumor mutational burden (TMB) have limited predictive value in this specific molecular context [12, 18, 24]. PD-L1 expression in SMARCA4-deficient tumors demonstrates heterogeneous distribution patterns and questionable clinical utility as a standalone biomarker [18]. While some studies have reported response to ICIs in patients with high PD-L1 expression [24, 25], others have observed limited efficacy even in PD-L1-positive cases [12]. This discrepancy suggests that additional factors beyond PD-L1 expression significantly influence treatment outcomes. Interestingly, one study found that PD-L1-positive patients with SMARCA4 mutations had significantly longer progression-free survival (8.3 vs. 5.1 months) and overall survival (18.9 vs. 9.3 months) when treated with chemoimmunotherapy [22].
Tumor mutational burden (TMB) shows limited standalone predictive value in this context. Although SMARCA4-deficient tumors may exhibit higher TMB [26], this does not consistently correlate with improved outcomes to immunotherapy [10, 22]. This dissociation is largely attributed to the countervailing immunosuppressive effects of frequent co-mutations such as STK11 and KEAP1, as discussed below. In fact, some studies have found no significant correlation between TMB levels and survival outcomes in SMARCA4-mutant patients receiving ICIs [22, 27]. This paradox may be explained by the counterbalancing effects of specific co-mutations that promote immune evasion despite high neoantigen load.

Genomic co-alterations: STK11, KEAP1, and KRAS
Specific genomic co-alterations critically determine both inherent tumor aggressiveness (a prognostic effect) and immunotherapy efficacy (a predictive effect). STK11 and KEAP1 mutations, in particular, are strongly associated with resistance to ICIs in SMARCA4-deficient tumors [10, 22]. It is important to distinguish that this association reflects both a predictive effect (worse outcome specifically with immunotherapy) and a broader prognostic effect (inferior overall survival irrespective of therapy). Patients with concurrent STK11 or KEAP1 mutations demonstrate significantly shorter progression-free survival and overall survival when treated with immunotherapy compared to those without these alterations [10, 22, 28]. This negative predictive impact is especially pronounced in the KRAS-mutant subgroup [29]. However, multivariate analyses have confirmed that STK11 mutations significantly increase the risk of death regardless of treatment modality, establishing a strong prognostic role [21, 28]. This suggests that the poor outcomes associated with STK11 mutations in patients receiving immunotherapy are due in part to an inherently more aggressive tumor biology, rather than being solely a reflection of immunotherapy resistance.

Emerging and exploratory biomarker approaches
Beyond conventional markers and co-mutations, several emerging biomarker approaches are being explored to improve patient selection. The specific class of SMARCA4 mutation may serve as a predictive biomarker for response to immunotherapy. SMARCA4 mutations are broadly categorized into two classes based on their predicted impact on the BRG1 protein [6, 19]. Class I alterations (including frameshift, nonsense, and splice-site mutations, as well as gene fusions and homozygous deletions) are predicted to result in a complete loss of functional protein, which is consistently confirmed by loss of BRG1 immunostaining [6]. In contrast, Class II alterations (primarily missense mutations) may allow for the production of a full-length but dysfunctional protein, with BRG1 expression often retained albeit potentially abnormal. This molecular distinction has clinical relevance. Class I alterations, representing a more profound loss of SWI/SNF complex function, are associated with a worse prognosis overall. Intriguingly, some evidence suggests they may also predict a relatively better response to immune checkpoint inhibitors compared to Class II mutations [6]. The proposed mechanism is that complete antigen loss (neoantigens from truncating mutations) or the genomic instability associated with profound chromatin remodeling defects in Class I tumors could potentially enhance immunogenicity. This differential response underscores the importance of comprehensive molecular characterization that includes precise mutational classification alongside assessment of BRG1 protein expression status. Additionally, the expression of cancer-testis antigens such as MAGEA4, CT45A, and PRAME is enriched in SMARCA4-truncated NSCLCs and may represent both a predictive biomarker and a therapeutic target.
Novel computational approaches are also under investigation. Machine learning algorithms have identified mutation signatures that predict response to anti-PD-1 therapy with higher accuracy than conventional biomarkers [30, 31]. These signatures include genes involved in various cellular processes and may provide a more comprehensive assessment of the complex tumor-immune interactions that determine treatment efficacy [31].

Clinical outcomes and efficacy of immune checkpoint inhibitors
The clinical efficacy of immune checkpoint inhibitors in SMARCA4-deficient thoracic tumors has yielded mixed results, as reported predominantly in retrospective cohort studies, with evidence of both remarkable responses in a subset of patients and general resistance in the majority. Understanding the factors that distinguish responders from non-responders is crucial for optimizing patient selection and treatment strategies.
Large retrospective cohorts consistently show that SMARCA4 alterations are linked to poorer outcomes after immunotherapy [6, 10, 29]. Specifically, in ICI-treated NSCLC, SMARCA4-mutant tumors have significantly worse objective response rates (ORR), progression-free survival (PFS), and overall survival (OS) than SMARCA4-wild type tumors. This detrimental effect is particularly pronounced in patients with concurrent KRAS mutations, where SMARCA4 alterations confer dramatically worse outcomes to immunotherapy. The median overall survival for SMARCA4-mutant patients receiving ICIs ranges from 8.2 to 9.5 months in different studies, compared to 13.1–15.5 months for SMARCA4-wild type patients [6, 29].
Despite this overall trend of resistance, data from small retrospective series and case reports indicate that a distinct subset of patients with SMARCA4-deficient tumors experience durable responses to immunotherapy [18, 24, 25]. These exceptional responders highlight the clinical and biological heterogeneity within SMARCA4-deficient tumors. Even in the highly aggressive SMARCA4-UT, which exemplifies the most challenging end of the clinical spectrum, durable responses to immunotherapy have been documented, as in a series where five patients had responses lasting ≥ 2 years [18]. This demonstrates that the profound immunosuppression driven by SMARCA4 deficiency is not absolute and underscores the critical need to identify the determinants of response within the larger SMARCA4-dNSCLC population. Similarly, another study in a specific patient cohort reported an ORR of 41.7% in SMARCA4-deficient thoracic tumors treated with PD-1/PD-L1 inhibitors, with all responders receiving ICIs as first-line therapy. This relatively high response rate likely reflects the selection of patients with more favorable baseline characteristics, rather than contradicting the general pattern of immunotherapy resistance in this tumor type. In these responding subsets, responses occurred across a range of PD-L1 expression levels (from < 1% to 100%), suggesting that factors beyond PD-L1 contribute to treatment sensitivity in this minority of patients [24].
The combination of immunotherapy with chemotherapy (chemoimmunotherapy) appears to provide superior outcomes compared to either modality alone in SMARCA4-deficient tumors [22, 27]. In patients with metastatic SMARCA4-dNSCLC, the addition of ICIs to chemotherapy significantly improved both median PFS (8.9 vs. 4.2 months) and median OS (19.7 vs. 13.1 months) compared to non-ICI treatments [27]. Similarly, immune-based therapy significantly improved median PFS (8.6 vs. 3.3 months) and OS (13.1 vs. 8.0 months) compared to non-immunotherapy in a comprehensive analysis of SMARCA4-deficient thoracic tumors [4].
A limited number of case reports have documented exceptional responses to combination approaches incorporating immunotherapy with other modalities [25, 32, 33], though these findings await validation in larger prospective studies. One patient with advanced SMARCA4-UT achieved a partial response lasting 31 months and progression-free survival of 43 months with a combination of pembrolizumab and anlotinib (a multi-targeted anti-angiogenic agent) [25]. Another patient with SMARCA4-dNSCLC and AIDS achieved overall survival exceeding 22 months with first-line immunotherapy plus anti-angiogenic therapy [33]. These case studies suggest that strategic combination therapies may help overcome the inherent resistance mechanisms in SMARCA4-deficient tumors.
The timing of immunotherapy initiation may also influence outcomes. In locally advanced SMARCA4-dNSCLC, the addition of ICIs to local treatment significantly improved median PFS (15.0 vs. 7.7 months) compared to non-ICI treatments, though no significant difference in OS was observed [27]. For stage IV patients, the addition of local treatment to the primary lesion did not significantly affect outcomes among ICI-treated patients, suggesting that systemic therapy with ICIs should be the primary focus in metastatic disease.

Emerging therapeutic strategies and novel targets beyond immunotherapy
Given the limitations of current immunotherapeutic approaches in SMARCA4-deficient thoracic tumors, significant research efforts are focused on developing novel strategies that target the unique vulnerabilities of these malignancies. Several promising therapeutic avenues are emerging that may complement or surpass the efficacy of existing immunotherapies.
One preclinically promising approach involves targeting the ferroptosis pathway. Preclinical studies have revealed that SMARCA4-deficient NSCLC cells demonstrate heightened susceptibility to ferroptosis, an iron-dependent form of cell death [34]. Mechanistically, SMARCA4 promotes chromatin accessibility and expression of ALDH16A1, which facilitates the translocation of thioredoxin (TXN) to lysosomes for degradation and directly inhibits TXN’s oxidoreductase function. Restoring ALDH16A1 levels or inhibiting TXN significantly enhances the effectiveness of chemo/immunotherapy in a ferroptosis-dependent manner, providing a strong rationale for combining ferroptosis inducers with existing therapies.
Epigenetic therapies represent another promising strategy for SMARCA4-deficient tumors. The synthetic lethality between SMARCA4 and SMARCA2, the alternative catalytic subunit of the SWI/SNF complex, has generated interest in targeting SMARCA2 or related epigenetic regulators. In preclinical models, EZH2 inhibitors have shown efficacy in SMARCA4-deficient contexts by counteracting the PRC2-mediated repression that occurs in the absence of functional SWI/SNF complex [3, 8]. Additionally, histone deacetylase inhibitors (HDACi) and other chromatin-modifying drugs may reverse the epigenetic alterations that drive immune evasion and treatment resistance [8, 35].
Cell cycle targeting through CDK4/6 inhibition has demonstrated potential in preclinical models of SMARCA4-deficient tumors [3]. The loss of SMARCA4 function creates dependencies on specific cell cycle regulators that can be therapeutically exploited. Similarly, inhibitors of ATR, AURKA, and other kinases involved in DNA damage response and cell division show promise in targeting the vulnerabilities associated with SMARCA4 deficiency.
Oxidative phosphorylation (OXPHOS) inhibition represents another novel approach for targeting SMARCA4-deficient tumors [16]. Preclinical studies have indicated that these tumors may be particularly dependent on mitochondrial metabolism, providing a therapeutic window for OXPHOS inhibitors. This metabolic vulnerability could be leveraged in combination with other targeted therapies or immunotherapy to enhance antitumor efficacy.
Anti-angiogenic agents in combination with immunotherapy have shown preliminary promising results in isolated case reports and small case series [25, 32], suggesting a potential therapeutic avenue that requires further investigation in controlled trials. The addition of drugs such as fruquintinib or anlotinib to PD-1/PD-L1 inhibitors may help normalize the tumor vasculature and improve immune cell infiltration into the tumor microenvironment. This approach may be particularly relevant for SMARCA4-deficient tumors, which often exhibit an immune-excluded or immune-desert phenotype.
Emerging biomarkers such as MTAP deficiency may also inform novel therapeutic strategies [36]. MTAP loss occurs in approximately 13% of NSCLCs and may co-occur with SMARCA4 deficiency, although this combination is not exceedingly common. This deficiency creates dependence on PRMT5, which can be targeted with specific inhibitors currently in clinical development. Therefore, while representing a subset, tumors harboring co-occurring MTAP and SMARCA4 deficiencies represent a biologically rational population for evaluating PRMT5-targeted therapies.
Furthermore, the well-characterized immunosuppressive TME of SMARCA4-deficient tumors, particularly those with concurrent KEAP1 or STK11 mutations, provides a rationale for exploring immune checkpoint blockade beyond PD-1/PD-L1 inhibition. Recent work has demonstrated that dual CTLA-4 and PD-1 blockade can overcome the immune resistance associated with KEAP1/STK11 co-mutated lung adenocarcinoma, leading to improved tumor control in preclinical models and promising clinical activity [37]. Given that these co-mutations frequently co-occur with SMARCA4 deficiency and drive a shared “immune-cold” phenotype, this combinatorial immunotherapy strategy represents a compelling and directly translatable therapeutic option for this challenging subgroup. Future clinical trials specifically stratifying for SMARCA4 status will be crucial to validate the efficacy of CTLA-4/PD-1 co-blockade in SMARCA4-deficient thoracic tumors.

Future perspectives
SMARCA4-deficient thoracic tumors represent a challenging subset of thoracic malignancies characterized by aggressive clinical behavior, distinct clinicopathological features, and complex molecular landscapes. The current review has comprehensively examined the multifaceted nature of these tumors, with particular emphasis on their interaction with the immune system and response to immunotherapeutic approaches.
The evidence to date clearly indicates that SMARCA4 deficiency confers generally poor prognosis and limited responsiveness to conventional immunotherapy. Importantly, the documented durable responses in a subset of patients [6, 18, 24] do not contradict this overarching trend but rather emphasize the existence of clinically and biologically distinct subgroups within this aggressive disease. This heterogeneity in treatment response underscores the importance of developing more sophisticated biomarkers that can disentangle predictive from prognostic effects. Future models should incorporate not only PD-L1 expression and tumor mutational burden, but also specific mutation patterns (such as STK11/KEAP1 co-mutations) with careful interpretation of their respective predictive and prognostic contributions, SMARCA4 alteration classes, and comprehensive assessments of the tumor immune microenvironment [12, 19, 22]. Exploratory studies using machine learning approaches to develop multidimensional predictive signatures show promise in this regard [30, 31], though their clinical utility remains to be prospectively validated.
From a biological perspective, the previously described profoundly immunosuppressive tumor microenvironment represents a major therapeutic vulnerability [7, 11, 23]. Strategies aimed at reversing the epigenetic silencing of immunostimulatory genes, inducing ferroptosis, or normalizing the tumor vasculature may help convert immunologically “cold” tumors into “hot” ones that are more susceptible to immune attack [8, 25, 34].
Looking forward, several key areas warrant focused research attention. First, prospective clinical trials specifically designed for patients with SMARCA4-deficient thoracic tumors are urgently needed to establish standardized treatment approaches. These trials should incorporate comprehensive biomarker assessments to identify patient subsets most likely to benefit from specific therapeutic combinations. Second, the development of novel therapeutic strategies that target the unique vulnerabilities of SMARCA4-deficient cells, such as synthetic lethal interactions with SMARCA2, EZH2, or PRMT5, should be prioritized. Several of these strategies have now entered early-phase clinical evaluation. For instance, EZH2 inhibitors (such as tazemetostat) and PRMT5 inhibitors are being investigated in early-stage trials encompassing tumors with SWI/SNF alterations, offering a direct path to clinical translation for SMARCA4-deficient thoracic malignancies [38]. Similarly, combination strategies involving immune checkpoint inhibitors with anti-angiogenic agents or novel metabolic inhibitors are also undergoing evaluation in phase I/II oncology trials, though their specific efficacy in this molecular subset awaits dedicated reporting. Finally, greater understanding of the cellular origins and evolutionary trajectories of these tumors may reveal additional therapeutic opportunities. Additionally, elucidating the impact of SMARCA4 deficiency on the efficacy of established and emerging targeted therapies will be crucial for optimizing personalized treatment sequences and combination strategies.
In conclusion, while SMARCA4-deficient thoracic tumors present significant therapeutic challenges, recent advances in understanding their biology and immune interactions provide a foundation for developing more effective treatment strategies. The advancement of several mechanism-based strategies into early clinical testing is a pivotal development. Through continued research efforts focused on overcoming mechanisms of immune evasion and identifying novel therapeutic vulnerabilities, there is hope for improving outcomes for patients with these aggressive malignancies. The integration of multimodal approaches combining immunotherapy with targeted agents, epigenetic modulators, and other novel therapeutics represents the most promising path forward in the management of this difficult-to-treat disease.