Bintrafusp Alfa (M7824): A Comprehensive Review of a First-in-Class Bifunctional Immunotherapy from Bench to Bedside and Beyond

Executive Summary

Bintrafusp alfa, also known as M7824, represents a landmark case study in modern immuno-oncology drug development. As a pioneering, first-in-class bifunctional fusion protein, it was engineered with a compelling scientific rationale: to simultaneously inhibit two key, non-redundant immunosuppressive pathways within the tumor microenvironment (TME). The molecule combines a Programmed Death-Ligand 1 (PD-L1) targeting antibody with a "trap" for Transforming Growth Factor-β (TGF-β), aiming to achieve a synergistic anti-tumor effect superior to that of single-pathway agents. This innovative approach was supported by a robust preclinical data package demonstrating enhanced tumor control, favorable immune modulation, and reversal of tumor cell plasticity.

This early promise led to the ambitious, multi-tumor INTR@PID global clinical trial program and a landmark strategic alliance between its discoverer, Merck KGaA, Darmstadt, Germany, and GlaxoSmithKline (GSK), valued at up to $4.2 billion. However, the trajectory of bintrafusp alfa serves as a cautionary tale. The program faced high-profile, definitive failures in two pivotal trials. The INTR@PID Lung 037 study, a head-to-head comparison against the standard-of-care pembrolizumab in first-line non-small cell lung cancer (NSCLC), was discontinued for futility, with bintrafusp alfa showing no superior efficacy and a significantly less favorable safety profile. Subsequently, the INTR@PID BTC 047 study in second-line biliary tract cancer (BTC) failed to meet its primary endpoint for regulatory submission, despite showing some activity in this difficult-to-treat disease.

These setbacks ultimately led to the mutual termination of the Merck-GSK alliance in September 2021, with no clinical milestone payments having been made beyond the initial upfront fee. In contrast to these failures, bintrafusp alfa demonstrated more encouraging and durable clinical activity in certain niche populations, particularly in human papillomavirus (HPV)-associated malignancies like cervical and head and neck cancers, where TGF-β biology is known to be dysregulated. Its safety profile, while generally described as manageable, proved complex, combining the known spectrum of immune-related adverse events from PD-L1 blockade with unique toxicities, such as skin lesions, attributed to TGF-β inhibition.

The story of bintrafusp alfa offers critical lessons for the field. It highlights the inherent risks of fixed-dose combination strategies within a single molecule, which sacrifices dosing flexibility. It underscores the paramount importance of a robust, prospectively validated biomarker strategy for targeted immunotherapies, as the drug's efficacy appears highly context-dependent on the specific tumor biology. Ultimately, the bintrafusp alfa program provides invaluable insights into the complexities of targeting TGF-β and informs the future design and clinical development strategy for the next generation of bifunctional antibodies and combination immunotherapies in oncology.

Molecular Design and Dual-Targeting Rationale

The development of bintrafusp alfa was predicated on an advanced understanding of tumor immunology, aiming to overcome the limitations of first-generation checkpoint inhibitors by simultaneously targeting two distinct mechanisms of immune evasion. Its molecular architecture and the scientific hypothesis underpinning its design represent a significant step in the evolution of cancer therapeutics.

Engineering a First-in-Class Bifunctional Fusion Protein

Bintrafusp alfa (DrugBank ID: DB15387), also known by its development codes M7824 and MSB0011359C, is a complex biologic drug classified as a biotech product, specifically a recombinant fusion protein and monoclonal antibody.[1] Its unique identity is confirmed by its CAS Number, 1918149-01-5.[2]

The molecule is an elegantly engineered bifunctional agent. Its structure consists of a human immunoglobulin G1 (IgG1) monoclonal antibody that specifically targets and blocks PD-L1.[2] This antibody component, which has a lambda light chain, is reported to be similar to the approved anti-PD-L1 antibody avelumab.[2] Covalently attached to the C-terminus of each of the antibody's two heavy chains, via a flexible peptide linker, is the soluble extracellular domain of the human TGF-β receptor II (TGF-βRII).[2] This configuration creates a potent "TGF-β trap" designed to bind and neutralize all three major isoforms of the cytokine: TGF-β1, TGF-β2, and TGF-β3.[7] The entire construct is of human sequence origin, minimizing potential immunogenicity.[10] The detailed amino acid sequences of the heavy chain (including the linker and TGF-βRII domain) and the light chain have been fully characterized and are publicly accessible.[10]

This molecular design was a deliberate and high-risk strategic choice. By physically linking the two functional domains into a single molecule, the developers aimed to achieve more than just the additive effect of two separate drugs. The core premise was that the anti-PD-L1 component would guide the molecule to the tumor microenvironment, where PD-L1 is often expressed, thereby concentrating the TGF-β trap precisely where it is needed most. This concept of "co-localized inhibition" was hypothesized to be more effective at remodeling the immunosuppressive TME than the systemic administration of separate anti-PD-L1 and anti-TGF-β agents. However, this design inherently creates a fixed-ratio therapeutic. The inability to independently adjust the dose of the PD-L1-blocking component versus the TGF-β-trapping component proved to be a critical vulnerability, as the optimal biological dose and toxicity profile for each pathway may differ significantly.

Table 1: Key Properties of Bintrafusp Alfa

Property	Description
Drug Name(s)	Bintrafusp alfa (English), M7824, MSB0011359C 2
DrugBank ID	DB15387 11
Type	Biotech 1
CAS Number	1918149-01-5 2
International Nonproprietary Name (INN)	bintrafusp alfa (Number: 10665) 2
Molecular Class	Recombinant Fusion Protein, Monoclonal Antibody, Antibody-dependent cell cytotoxicity stimulant, Cytotoxic T lymphocyte stimulant, Natural killer cell stimulant 2
Isotype	Human IgG1 lambda 6
Targets	Programmed cell death-1 ligand-1 (PD-L1), Transforming growth factor beta (TGF-β) 3
Developer(s)	Discovered in-house at Merck KGaA, Darmstadt, Germany (EMD Serono); formerly in a strategic alliance with GSK 13

The Scientific Imperative for Co-localized Inhibition

The rationale for creating bintrafusp alfa stems from the recognition that tumors employ multiple, often synergistic, pathways to evade immune destruction. Targeting a single pathway, such as the PD-1/PD-L1 axis, has revolutionized cancer treatment but is effective in only a subset of patients, suggesting that other resistance mechanisms are at play.[7] The TGF-β and PD-L1 pathways were identified as key complementary and non-redundant drivers of immunosuppression, making their simultaneous blockade a highly rational therapeutic strategy.[7]

The Role of PD-L1: The interaction between PD-1 on activated T-cells and its ligand, PD-L1, on tumor cells and other cells in the TME, acts as a crucial immune checkpoint. This binding delivers an inhibitory signal that dampens T-cell and Natural Killer (NK) cell effector functions, thereby preventing them from attacking the tumor.[7] Monoclonal antibodies that block this axis have demonstrated unprecedented and durable clinical benefits across a wide range of malignancies, validating it as a central pillar of modern immuno-oncology.[7]

The Multifaceted Role of TGF-β in Cancer: TGF-β is a pleiotropic cytokine with a complex, context-dependent role in cancer.[7] In normal tissues and early-stage carcinogenesis, it can act as a tumor suppressor by inhibiting the cell cycle and promoting apoptosis.[17] However, in established and advanced cancers, its function paradoxically switches to that of a potent tumor promoter.[8] Dysregulated TGF-β signaling in the TME promotes cancer progression through several interconnected mechanisms:

• Profound Immunosuppression: TGF-β is a master regulator of immune suppression. It directly inhibits the function of cytotoxic T-lymphocytes, helper T-cells, and NK cells. It also promotes the expansion and function of immunosuppressive cell populations, such as regulatory T-cells (Tregs) and M2-polarized macrophages, effectively shielding the tumor from immune attack.[7]
• Induction of Epithelial-to-Mesenchymal Transition (EMT): TGF-β is a primary driver of EMT, a cellular reprogramming process in which cancer cells lose their organized, epithelial characteristics and acquire migratory, invasive, mesenchymal features. This transition is strongly associated with metastasis, tumor progression, and resistance to both chemotherapy and immunotherapy.[7]
• TME Remodeling: TGF-β promotes a tumor-friendly microenvironment by stimulating angiogenesis (the formation of new blood vessels) and fibrosis (the deposition of extracellular matrix), which can create a physical barrier that prevents immune cell infiltration.[7]
• Cross-talk with the PD-L1 Pathway: Importantly, preclinical research demonstrated a direct mechanistic link between the two pathways. In NSCLC cell lines, TGF-β signaling was shown to induce the transcription and protein expression of PD-L1, suggesting that high levels of TGF-β in the TME can directly contribute to resistance to anti-PD-1/PD-L1 therapies by upregulating the target.[7]

The entire premise of bintrafusp alfa was built on the hypothesis that a single molecule capable of co-localized targeting would be superior to simply combining two separate antibodies. Preclinical studies directly tested this, showing that bintrafusp alfa elicited distinct and superior anti-tumor responses compared to the co-administration of an anti-PD-L1 antibody and a separate TGF-β trap.[19] The failure of this elegant and scientifically validated concept to translate into superior clinical outcomes in broad patient populations remains the central paradox of the bintrafusp alfa story, pointing to the profound gap that can exist between preclinical models and the complexities of human tumor biology and inter-patient heterogeneity.

Preclinical Evidence and Proof-of-Concept

Before advancing into large-scale human trials, bintrafusp alfa was subjected to a rigorous preclinical evaluation that provided a strong foundation for its clinical development. A key advantage of the molecule was its ability to bind to both human and murine versions of PD-L1 and TGF-β, which allowed for comprehensive testing in clinically relevant syngeneic mouse models as well as with human cancer cells in vitro.[7] The results from these studies were compelling and consistently supported the hypothesis of synergistic anti-tumor activity.

In in vivo studies using the MC38 colorectal cancer syngeneic mouse model, bintrafusp alfa monotherapy demonstrated significantly greater reductions in tumor volume compared to treatment with either a standalone TGF-β trap or an anti-PD-L1 antibody alone.[8] This provided direct evidence for the superiority of the dual-targeting approach over inhibiting either pathway individually. Mechanistically, this enhanced anti-tumor effect was associated with profound changes within the tumor microenvironment. Analysis of treated tumors revealed that bintrafusp alfa led to a substantial increase in the infiltration of beneficial immune effector cells, including CD8+ cytotoxic T-lymphocytes and NK cells, while simultaneously decreasing the infiltration of immunosuppressive myeloid-derived suppressor cells (MDSCs).[6]

Beyond general immune activation, preclinical studies demonstrated that bintrafusp alfa could directly counteract key tumor-promoting processes driven by TGF-β. In vitro experiments using human lung cancer cell lines showed that bintrafusp alfa could both prevent the initiation of and reverse established TGF-β-induced EMT.[7] Treatment with the fusion protein prevented the loss of epithelial markers like E-cadherin and blocked the upregulation of mesenchymal markers such as vimentin and Snail.[7] This anti-EMT effect, which was attributed specifically to the TGF-β trap component of the molecule, was also shown to overcome tumor cell resistance to chemotherapy.[7]

Furthermore, bintrafusp alfa was shown to enhance the immune system's ability to directly kill cancer cells. The molecule was capable of mediating antibody-dependent cell-mediated cytotoxicity (ADCC)—a process where immune cells like NK cells recognize and kill antibody-coated target cells—against a range of human cancer cell lines, including those from bladder, cervical, and triple-negative breast cancers.[7] This activity was dependent on the molecule's ability to bind to PD-L1 on the tumor cell surface.[7] In a complex but important finding, it was observed that TGF-β-induced upregulation of PD-L1 on NSCLC cells actually rendered them

more sensitive to ADCC mediated by bintrafusp alfa, suggesting an intricate interplay between the two targeted pathways.[7]

This exceptionally strong preclinical data package, demonstrating superiority over monotherapies, favorable TME modulation, and direct effects on tumor cell plasticity, formed the scientific bedrock for the subsequent clinical program and the major corporate investment from GSK. The stark contrast between this robust preclinical promise and the eventual clinical outcomes in pivotal trials highlights the persistent and formidable challenge of translating findings from simplified model systems to the heterogeneous and complex reality of human cancer. In retrospect, the preclinical emphasis on TGF-β's role in EMT and immune exclusion might have hinted that the drug's success would be confined to specific tumor subtypes defined by these characteristics, rather than the broader, PD-L1-defined populations that were initially pursued in the clinic.

The INTR@PID Clinical Development Program: A Multi-Tumor Assessment

The INTR@PID (Investigational Trial Program in Difficult-to-Treat Cancers) clinical program was a vast and ambitious global effort designed to evaluate the efficacy and safety of bintrafusp alfa across a wide spectrum of malignancies, particularly those where TGF-β was hypothesized to be a key driver of disease progression and immunosuppression.[14] The program ultimately enrolled over 1,300 patients with various solid tumors, reflecting the high hopes placed on this novel bifunctional agent.[15]

Table 2: Summary of Major Clinical Trials in the INTR@PID Program

Trial Identifier	Phase	Indication	Patient Population	Intervention(s)	Status/Outcome
INTR@PID Lung 037 (NCT03631706)	III	Non-Small Cell Lung Cancer (NSCLC)	1L, PD-L1 high	Bintrafusp alfa vs. Pembrolizumab	Terminated for futility 21
INTR@PID BTC 047 (NCT03833661)	II	Biliary Tract Cancer (BTC)	2L, post-platinum	Bintrafusp alfa monotherapy	Completed; failed to meet primary endpoint for filing 1
INTR@PID 001 (NCT02517398)	I	Advanced Solid Tumors	Heavily pretreated	Bintrafusp alfa dose escalation/expansion	Completed; provided early signals in various tumors (SCCHN, HPV-cancers) 8
NCT04551950	Ib	Cervical Cancer	1L (P/R/M or locally advanced)	Bintrafusp alfa + Chemo +/- Bevacizumab	Completed; showed promising activity 26
BIMES (NCT05005429)	II	Malignant Pleural Mesothelioma	Previously treated	Bintrafusp alfa monotherapy	Completed 27
NCT04874311	II	Advanced Soft Tissue Sarcoma	Advanced/Metastatic	Bintrafusp alfa + Doxorubicin	Recruiting 28
NCT03554473	I	Small Cell Lung Cancer (SCLC)	Relapsed	Bintrafusp alfa + Topotecan or Temozolomide	Completed 29
NCT02723955	I	Advanced Solid Tumors	Selected advanced tumors	Bintrafusp alfa + other agents	Completed 11
NCT05061823	N/A	Rollover Study	Participants from parent studies	Bintrafusp alfa	Active, not recruiting 30

The Litmus Test: First-Line Non-Small Cell Lung Cancer (INTR@PID Lung 037)

The centerpiece of the INTR@PID program was the INTR@PID Lung 037 trial (NCT03631706). This was not a cautious exploratory study but a bold, adaptive Phase III trial designed to challenge the reigning standard of care.[22] The study randomized patients with previously untreated, advanced NSCLC with high PD-L1 expression to receive either bintrafusp alfa or the market-leading anti-PD-1 antibody, pembrolizumab.[23] The strategic goal was unambiguous: to prove that the dual-targeting mechanism of bintrafusp alfa was superior to single-pathway PD-1 blockade in this key patient population.[34]

This high-risk, high-reward strategy ended in unequivocal failure. In January 2021, Merck KGaA and GSK announced that the trial was being discontinued on the recommendation of the Independent Data Monitoring Committee, which concluded that the study was unlikely to meet its co-primary endpoint of progression-free survival (PFS).[2]

The final published results from the 304 randomized patients confirmed the futility of the trial, revealing that bintrafusp alfa was not only non-superior but was numerically inferior to pembrolizumab on key efficacy metrics while being substantially more toxic.[23] This outcome was a critical blow, as success in first-line NSCLC is often a gateway to blockbuster status. The failure demonstrated that in a broad population selected only by PD-L1 status, the hypothesized synergistic benefit of adding TGF-β trapping was insufficient to outperform the established efficacy of pembrolizumab alone.

Table 3: Comparative Efficacy of Bintrafusp Alfa vs. Pembrolizumab in INTR@PID Lung 037 (NCT03631706)

Endpoint	Bintrafusp Alfa Arm (n=152)	Pembrolizumab Arm (n=152)	Hazard Ratio (95% CI) / p-value
Median PFS (IRC)	7.0 months	11.1 months	1.232 (0.885–1.714) 23
Median OS	21.1 months	22.1 months	1.201 (0.796–1.811) 23
ORR (unconfirmed)	46.7%	51.3%	p = 0.401 35
Median DoR	Not Reached	Not Reached	N/A 35

A Challenging Indication: Second-Line Biliary Tract Cancer (INTR@PID BTC 047)

Following the NSCLC failure, hopes pivoted to other indications. The INTR@PID BTC 047 trial (NCT03833661) was a Phase II, single-arm study that enrolled 159 patients with locally advanced or metastatic BTC whose disease had progressed after first-line platinum-based chemotherapy.[15] This is a patient population with a grim prognosis and very limited effective treatment options, a setting where even modest activity can be considered clinically meaningful.[15]

In March 2021, the companies announced that this trial had also failed to meet its primary objective. While bintrafusp alfa demonstrated single-agent activity, the observed objective response rate (ORR) did not cross the pre-defined threshold that would have been required to support a regulatory filing.[15]

The final published data showed an Independent Review Committee (IRC)-adjudicated ORR of 10.7%.[24] While this was numerically higher than the historical ORRs of 5-8% seen with other immunotherapies in this setting, it was not deemed sufficient for a breakthrough designation.[34] A notable finding was the durability of responses in the small subset of patients who did respond, with a median duration of response (DoR) of 10.0 months.[24] However, this was overshadowed by a very short median PFS of just 1.8 months, indicating that the majority of patients did not benefit.[24]

Table 4: Efficacy and Survival Outcomes in the INTR@PID BTC 047 Trial (NCT03833661)

Metric	Result (Value, 95% CI)
ORR (IRC)	10.7% (6.4–16.6%) 38
Complete Response Rate	1.9% 38
Partial Response Rate	8.8% 38
Median DoR	10.0 months (1.9–15.7) 38
Median PFS (IRC)	1.8 months (1.7–1.8) 38
Median OS	7.6 months (5.8–9.7) 38
12-month OS Rate	38.8% 38
18-month OS Rate	26.9% 38

A Glimmer of Hope: HPV-Associated Malignancies

In stark contrast to the disappointing results in NSCLC and BTC, bintrafusp alfa showed consistent and promising signals of activity in tumors driven by HPV. This finding is biologically plausible, as dysregulation of the TGF-β signaling pathway is a known consequence of HPV infection, potentially making these tumors uniquely susceptible to bintrafusp alfa's mechanism.[39]

• Cervical Cancer: A pooled analysis of 39 heavily pretreated patients with recurrent/metastatic cervical cancer from two studies (NCT02517398 and NCT03427411) revealed an encouraging ORR of 28.2% and a median OS of 13.4 months.[41] More impressively, a Phase Ib trial (NCT04551950) combining bintrafusp alfa with standard chemotherapy +/- bevacizumab in first-line cervical cancer reported high ORRs of 75.0% in the triplet cohort and 44.4% in the doublet cohort, supporting further investigation.[26]
• Head and Neck Cancer (SCCHN): In a Phase I expansion cohort of 32 heavily pretreated SCCHN patients, the overall ORR was modest at 13%. However, the responses were remarkably durable, with a median DoR of 21.4 months and a 3-year OS rate of 24.0% after long-term follow-up.[25] Critically, the clinical activity was concentrated in the HPV-positive subgroup, which had an ORR of 33%, compared to just 5% in the HPV-negative subgroup.[39]
• Pooled Analysis: A comprehensive pooled analysis of 75 patients across various pretreated HPV-associated malignancies confirmed these findings, showing an overall ORR of 28.0% (including four complete responses) and a median OS of 21.3 months, all with a manageable safety profile.[42]

The divergent outcomes between the broad indication trials and the HPV-associated cancer cohorts represent the most critical lesson from the entire clinical program. It strongly suggests a failure in the initial clinical and biomarker strategy. The drug appears to have a potent, biologically relevant mechanism, but its efficacy is likely confined to tumors where the TGF-β pathway is a dominant driver of the immunosuppressive phenotype. The initial strategy of targeting a broad population based on a PD-L1 biomarker alone, rather than a TGF-β-related biomarker, was likely the program's fatal flaw.

Exploration Across Other Solid Tumors

The INTR@PID program's breadth is evident in the numerous other trials initiated. Completed Phase 1 trials explored bintrafusp alfa in diverse settings, including general solid tumors [11], prostate cancer [12], small cell lung cancer [29], and breast cancer.[44] Further Phase 2 investigations were launched in advanced soft tissue sarcoma (in combination with doxorubicin) and malignant pleural mesothelioma, indicating a persistent effort to find a responsive niche for the drug even after the major setbacks.[27]

Consolidated Safety and Tolerability Analysis

The safety profile of bintrafusp alfa is a direct reflection of its dual mechanism of action, combining the known toxicities of PD-L1 inhibition with a distinct set of adverse events attributable to the systemic blockade of TGF-β. While often described in clinical reports as "manageable," the cumulative data reveal a more complex and challenging profile compared to single-agent PD-(L)1 inhibitors, a factor that critically influenced its overall risk-benefit assessment.[8] The drug was administered to over 1,300 patients in the INTR@PID program, with the recommended Phase 2 dose (RP2D) established as 1200 mg intravenously every 2 weeks or 2400 mg every 3 weeks.[15]

In the pivotal INTR@PID Lung 037 trial, the safety profile of bintrafusp alfa was a significant liability. The rate of Grade 3-4 treatment-related adverse events (TRAEs) was 42.4%, more than three times higher than the 13.2% observed in the pembrolizumab arm.[23] This led to a correspondingly high rate of treatment discontinuation due to TRAEs (25.8% for bintrafusp alfa vs. 6.6% for pembrolizumab), indicating that the toxicity was a major clinical challenge.[35] In the context of no superior efficacy, this unfavorable safety profile created an unacceptable risk-benefit ratio.

In other settings, such as the Phase II trial in second-line biliary tract cancer, the rate of Grade ≥3 TRAEs was lower at 26.4%.[24] This demonstrates that the tolerability of a drug is not an absolute measure but is highly dependent on the patient population and disease context. In a setting with few effective options, a higher level of toxicity may be deemed acceptable if accompanied by meaningful and durable clinical benefit.

Adverse Events of Special Interest:

• Immune-Related Adverse Events (irAEs): As expected from its anti-PD-L1 component, bintrafusp alfa was associated with a spectrum of irAEs consistent with the class, including pruritus, rash, colitis, and pneumonitis.[13] In the BTC trial, irAEs occurred in 28.9% of patients, with 12.6% being Grade ≥3.[38]
• TGF-β-Related Skin Lesions: A unique and mechanistically important toxicity was the development of specific skin lesions attributed to TGF-β inhibition. These included keratoacanthoma, hyperkeratosis, and in some cases, cutaneous squamous cell carcinoma.[46] These events were noted in the INTR@PID LUNG 024 study as a key differentiator from standard immunotherapy toxicity and were observed in approximately 7% of patients in early pooled analyses.[46]
• Bleeding Events: Bleeding was monitored as an adverse event of special interest. In a trial in cervical cancer, bleeding events were common, though mostly low-grade.[26] However, Grade 3 gastrointestinal hemorrhages were reported in a study of a quadruple combination therapy involving bintrafusp alfa, highlighting a potential risk, especially when combined with other agents.[49]

Fatal adverse events were rare but did occur. One treatment-related death from hepatic failure was reported in the BTC trial, and three Grade 5 events (septic shock, interstitial lung disease) were seen in an earlier study in Asian BTC patients.[24]

An exposure-safety analysis conducted on data from 673 patients in early-phase trials found that drug exposure levels were generally weakly or not at all correlated with the probability of most adverse events, suggesting a relatively predictable safety profile that supported the selection of the RP2D.[46]

Table 5: Consolidated Safety Profile of Bintrafusp Alfa from Key Clinical Trials

Adverse Event	INTR@PID Lung 037 (NSCLC, 1L)	INTR@PID BTC 047 (BTC, 2L)	HPV-Associated Cancers (Pooled)
Any Grade TRAEs	82.1% 35	62.3% 38	84.6% 41
Grade ≥3 TRAEs	42.4% 23	26.4% 24	20.5% 41
Discontinuation due to TRAEs	25.8% 35	9.4% 38	N/A
Treatment-Related Deaths	0.7% 35	0.6% (1 patient) 24	0% 41
Common TRAEs (Any Grade)	Pruritus (31.8%), Rash (29.1%), Diarrhea (12.6%) 35	Pruritus (12.6%), Rash (9.4%), Fatigue (8.8%) 38	Pruritus (25.3%), Dermatitis Acneiform (24.0%), Anemia (18.7%) 42
TGF-β-Related Skin Lesions	Not specified, but higher AEs of special interest (53.0%) 35	8.2% 38	Keratoacanthoma (Grade 3, 1 patient) 41
Grade ≥3 irAEs	Higher than pembrolizumab (11.8%) 35	12.6% 38	Colitis, Pneumonitis (1 patient each) 41

The dual-target nature of bintrafusp alfa created a "double-edged sword" of toxicity. The drug carried the combined risks of both PD-L1 and TGF-β inhibition, resulting in a more complex and, in some settings, more severe adverse event profile than its single-target comparator. This inherent toxicity profile likely raised the efficacy bar for demonstrating a favorable risk-benefit balance and may have also posed a significant barrier to developing effective and tolerable combination regimens, limiting its potential therapeutic utility.

The Rise and Fall of a Blockbuster Alliance

The corporate history of bintrafusp alfa is as dramatic as its clinical story, marked by a blockbuster partnership that embodied the soaring optimism for next-generation immunotherapies, followed by a rapid dissolution in the face of disappointing clinical data.

Bintrafusp alfa was discovered and developed in-house by the scientists at Merck KGaA, Darmstadt, Germany, and its U.S. affiliate, EMD Serono.[15] The early clinical development program was driven by Merck KGaA, with numerous publications from Phase I trials featuring their researchers as key authors.[13]

The compelling preclinical data and promising early clinical signals culminated in a landmark event in February 2019. Merck KGaA announced a global strategic alliance with GlaxoSmithKline (GSK) to co-develop and co-commercialize bintrafusp alfa.[15] The deal was one of the largest biopharma licensing agreements of its time, with a total potential value of up to €3.7 billion (approximately $4.2 billion).[34] The financial structure of the agreement reflected the immense perceived potential of the asset. GSK made a significant upfront payment of €300 million, with the remainder of the value tied to the achievement of specific development and sales milestones, including up to €500 million for development milestones and a staggering €2.9 billion in potential sales-related payments.[34] This deal was not merely a bet on a single drug but an investment in a novel therapeutic platform—bifunctional fusion proteins—and a high-potential, unproven biological hypothesis.

However, the alliance was short-lived. The clinical failures of the INTR@PID program in 2021 triggered a rapid unraveling of the partnership. Following the definitive negative readouts from the pivotal trials in NSCLC and BTC, Merck KGaA and GSK announced a mutual decision to terminate their agreement, effective September 30, 2021.[14] The termination was explicitly and directly attributed to the clinical trial data, which failed to replicate the encouraging results seen in earlier, smaller studies.[14]

The financial fallout was stark and highlighted the binary nature of such high-risk R&D collaborations. Because bintrafusp alfa failed to meet the critical clinical endpoints that would trigger further payments, no milestone payments were ever made by GSK to Merck KGaA beyond the initial €300 million upfront investment.[14] The billions of dollars in potential value vanished. The rapid timeline—from a blockbuster deal in February 2019 to termination in September 2021—underscores the accelerated pace and high stakes of modern oncology development, where a single pivotal trial can create or destroy immense value almost overnight.

Following the termination, Merck KGaA retained the rights to bintrafusp alfa and stated its intention to continue analyzing the vast dataset from the INTR@PID program to deepen its scientific understanding of TGF-β biology.[14] While major company-sponsored development has ceased, some clinical investigations have continued under the sponsorship of academic institutions and cooperative groups like the National Cancer Institute (NCI), primarily exploring the drug's potential in niche indications.[1] A rollover study (NCT05061823) was also established to provide continued access to the drug for patients who had derived benefit in the parent trials, a testament to the fact that the drug was effective for a small subset of individuals.[30]

Critical Analysis and Future Directions

The bintrafusp alfa saga, from its ambitious conception to its clinical disappointments, offers a wealth of knowledge for the future of oncology drug development. A critical examination of its failures and isolated successes provides crucial lessons for the design of next-generation immunotherapies and the strategic approach to their clinical evaluation.

Lessons Learned from the Bintrafusp Alfa Program

The comprehensive failure of bintrafusp alfa in its pivotal trials can be attributed to a confluence of factors related to its design, the complexity of its targets, and the clinical strategy employed.

• The Challenge of Fixed-Dose Combination: A primary lesson learned relates to the inherent risk of a bifunctional fusion protein design. By locking the anti-PD-L1 and anti-TGF-β components into a single molecule, the design eliminated the ability to independently titrate the doses of each functional part. This lack of dosing flexibility may have been a fatal flaw, preventing the optimization of the therapeutic window.[56] It is plausible that the dose required for effective TGF-β trapping in the TME induced unacceptable toxicity, or that the optimal biological doses for the two pathways are simply different. A strategy using two separate, co-administered antibodies would have allowed for this crucial flexibility.
• The Imperative of Biomarker-Driven Patient Selection: The most striking finding from the clinical program is the heterogeneity of response. The failures in broad populations (NSCLC, BTC) contrasted with promising signals in a biologically defined subset (HPV-associated cancers) points directly to a failure of biomarker strategy.[25] The program relied on PD-L1 expression, a marker for the PD-L1 component, but lacked a validated predictive biomarker for the TGF-β component. Future development of agents targeting the TGF-β pathway will almost certainly require the prospective identification of patients whose tumors are demonstrably dependent on or suppressed by TGF-β signaling, using markers such as TME fibrosis, EMT signatures, or high TGF-β expression.[56]
• The Disconnect Between Preclinical Models and Human Complexity: The story of bintrafusp alfa is a classic, if painful, reminder of the limitations of preclinical models.[16] The strong, consistent preclinical data showing synergy and superiority did not translate to the complex and heterogeneous environment of human cancer. This disconnect underscores the need for more sophisticated preclinical models that can better recapitulate the human TME and immune system to improve the predictive value of early-stage research.
• Context is Key for Efficacy and Safety: The drug's clinical profile demonstrates that "efficacy" and "safety" are not absolute concepts. The modest activity and notable toxicity that were unacceptable in the first-line NSCLC setting (where a highly effective and less toxic standard of care exists) might have been viewed more favorably in a setting of high unmet need, such as heavily pretreated HPV-positive cancers.[35] This highlights that the viability of a new drug is always judged relative to the existing therapeutic landscape for a specific indication.

The Future of TGF-β Inhibition in Oncology

Despite the high-profile failures of bintrafusp alfa and other agents like Novartis's nisevokitug, the scientific rationale for targeting TGF-β in cancer remains compelling, particularly as a strategy to overcome resistance to checkpoint inhibitors.[18] The field, however, is clearly pivoting away from the bintrafusp alfa model and embracing new strategies informed by its shortcomings.

• A Shift in Strategy: The focus has shifted towards developing TGF-β inhibitors as standalone agents designed for flexible combination with PD-(L)1 inhibitors and other therapies. This approach allows for independent dose optimization and titration to manage toxicity and maximize efficacy.[56]
• Next-Generation Agents: Development is now concentrated on more selective inhibitors that target specific TGF-β isoforms (e.g., TGF-β1 and -β3) to potentially widen the therapeutic window, as well as small-molecule inhibitors of the TGF-β receptor (TGFβR1), such as Medpacto's vactosertib, which is now one of the most advanced assets in this class.[56]
• Biomarker Enrichment: Learning directly from the bintrafusp alfa experience, companies like Scholar Rock are pursuing a strict biomarker-driven approach for their anti-TGF-β1 antibody, SRK-181, enrolling patients based on markers of an immune-excluded TME and active TGF-β signaling (e.g., pSMAD2).[56]

The Outlook for Bifunctional Antibodies

The failure of a single high-profile agent has not halted progress in the broader field of bispecific and multispecific antibodies. This area of drug development remains one of the most dynamic and innovative in oncology, with numerous candidates advancing through clinical trials.[57] The FDA approval of amivantamab, a bispecific antibody targeting EGFR and MET for a specific subset of NSCLC patients, provides a clear example of how this platform can succeed when the right targets are chosen for a well-defined disease context.[58]

The bintrafusp alfa program serves as a critical guidepost for future endeavors in this space. Success will likely depend on several key factors:

1. Rational Target Pairing: Selecting targets with a strong, scientifically validated synergistic or complementary relationship within a specific cancer type.
2. Deep Understanding of the TME: Moving beyond systemic effects to understand how the engineered antibody will behave within the complex and heterogeneous tumor microenvironment.
3. Sophisticated Protein Engineering: Designing constructs with optimized binding affinities, valency, and pharmacokinetic properties to ensure an adequate therapeutic window.
4. Pragmatic Clinical Strategy: Employing robust biomarker strategies from the outset and avoiding high-risk, head-to-head trials against dominant standards of care unless there is an exceptionally strong signal of superiority in a clearly defined patient population.

In conclusion, while bintrafusp alfa did not fulfill its initial promise, the extensive research and clinical data generated have provided the scientific community with invaluable knowledge. Its story has profoundly shaped our understanding of TGF-β biology, highlighted the challenges and opportunities in developing multifunctional therapeutics, and will undoubtedly influence the design and strategy of cancer drug development for years to come.

Works cited

1. Bintrafusp alfa | MedPath, accessed August 11, 2025, https://trial.medpath.com/drug/811dd01ce9f000bc/bintrafusp-alfa?tab=clinical_trials&page=2
2. bintrafusp alfa | Ligand page | IUPHAR/BPS Guide to ..., accessed August 11, 2025, https://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=11376
3. Bintrafusp alfa - EMD Serono/Merck - AdisInsight - Springer, accessed August 11, 2025, https://adisinsight.springer.com/drugs/800043052
4. Bintrafusp alfa - Caltag Medsystems, accessed August 11, 2025, https://www.caltagmedsystems.co.uk/pricing_ordering/product_detail.php?CI_ID=1714725&group_1=All&group_2=All&supplier=All&clonality=All&host=All&species=All&applications=All®ulatory_status=All&searchwords=stem%20cell
5. Bintrafusp alfa - Nordic Biosite, accessed August 11, 2025, https://nordicbiosite.com/product/HY-P99480-10/Bintrafusp-alfa
6. Bintrafusp alfa (M7824) | Bifunctional Fusion Protein - MedchemExpress.com, accessed August 11, 2025, https://www.medchemexpress.com/bintrafusp-alfa.html
7. Preclinical and clinical studies of bintrafusp alfa, a novel bifunctional anti-PD-L1/TGFβRII agent: Current status, accessed August 11, 2025, https://pmc.ncbi.nlm.nih.gov/articles/PMC9335510/
8. Bintrafusp Alfa, a Bifunctional Fusion Protein Targeting TGF-β and PD-L1, in Second-Line Treatment of Patients With NSCLC: Results From an Expansion Cohort of a Phase 1 Trial - PMC - PubMed Central, accessed August 11, 2025, https://pmc.ncbi.nlm.nih.gov/articles/PMC8210474/
9. Dual targeting of TGF-β and PD-L1 via a bifunctional anti-PD-L1/TGF-βRII agent: status of preclinical and clinical advances, accessed August 11, 2025, https://jitc.bmj.com/content/8/1/e000433
10. BINTRAFUSP ALFA - GSRS, accessed August 11, 2025, https://gsrs.ncats.nih.gov/ginas/app/beta/substances/NW9K8C1JN3
11. Bintrafusp alfa Completed Phase 1 Trials for Neoplasms Treatment | DrugBank Online, accessed August 11, 2025, https://go.drugbank.com/drugs/DB15387/clinical_trials?conditions=DBCOND0027948&phase=1&purpose=treatment&status=completed
12. Bintrafusp alfa Completed Phase 1 Trials for Advanced Solid Tumors / Metastatic Prostate Cancer / Solid Tumors / Neoplasms of the Prostate Treatment - DrugBank, accessed August 11, 2025, https://go.drugbank.com/drugs/DB15387/clinical_trials?conditions=DBCOND0030110%2CDBCOND0033188%2CDBCOND0080047%2CDBCOND0029860&phase=1&purpose=treatment&status=completed
13. Bintrafusp Alfa, a Bifunctional Fusion Protein Targeting TGF-β and PD-L1, in Second-Line Treatment of Patients With NSCLC: Results From an Expansion Cohort of a Phase 1 Trial - PubMed, accessed August 11, 2025, https://pubmed.ncbi.nlm.nih.gov/32173464/
14. Merck Announces Mutual Decision to End Bintrafusp Alfa ..., accessed August 11, 2025, https://www.merckgroup.com/en/news/bintrafusp-alfa-agreement-with-gsk.html
15. 2021-03-16 Topline Data for Bintrafusp Alfa as Second-Line Monotherapy Treatment in Biliary Tract Cancer - EMD Serono, accessed August 11, 2025, https://www.emdserono.com/us-en/company/news/press-releases/topline-data-for-bintrafusp-alfa-as-second-line-monotherapy-treatment-in-biliary-tract-cancer-16-03-2021.html
16. Phase I Study of the Bifunctional Fusion Protein Bintrafusp Alfa in Asian Patients with Advanced Solid Tumors, Including a Hepatocellular Carcinoma Safety‐Assessment Cohort, accessed August 11, 2025, https://pmc.ncbi.nlm.nih.gov/articles/PMC7485354/
17. Mechanism of action of bintrafusp alfa, a first‐in‐class bifunctional... - ResearchGate, accessed August 11, 2025, https://www.researchgate.net/figure/Mechanism-of-action-of-bintrafusp-alfa-a-first-in-class-bifunctional-fusion-protein_fig3_356721755
18. Targeting TGF-β in Gastrointestinal Cancers: Is it a Viable Therapeutic Strategy?, accessed August 11, 2025, https://dailynews.ascopubs.org/do/targeting-tgf--x03b2-gastrointestinal-cancers-viable-therapeutic-strategy
19. Colocalized targeting of TGF-β and PD-L1 by bintrafusp alfa elicits distinct antitumor responses - PubMed, accessed August 11, 2025, https://pubmed.ncbi.nlm.nih.gov/35858707/
20. Colocalized targeting of TGF-β and PD-L1 by bintrafusp alfa elicits distinct antitumor responses | Journal for ImmunoTherapy of Cancer, accessed August 11, 2025, https://jitc.bmj.com/content/10/7/e004122
21. Update on the INTR@PID Clinical Program Including Lung 037 Study - Merck Group, accessed August 11, 2025, https://www.merckgroup.com/en/news/bintrafusp-alfa-037-update-20-01-2021.html
22. M7824 Versus Pembrolizumab as a First-line (1L) Treatment in Participants With Programmed Death-ligand 1 (PD-L1) Expressing Advanced Non-small Cell Lung Cancer (NSCLC) | ClinicalTrials.gov, accessed August 11, 2025, https://clinicaltrials.gov/study/NCT03631706
23. Bintrafusp Alfa Versus Pembrolizumab in Patients With Treatment-Naive, Programmed Death-Ligand 1-High Advanced NSCLC: A Randomized, Open-Label, Phase 3 Trial - PubMed, accessed August 11, 2025, https://pubmed.ncbi.nlm.nih.gov/37597750/
24. Phase 2 trial of bintrafusp alfa as second-line therapy for patients with locally advanced/metastatic biliary tract cancers - PubMed, accessed August 11, 2025, https://pubmed.ncbi.nlm.nih.gov/36999533/
25. Long-term follow-up of bintrafusp alfa, a bifunctional fusion protein targeting TGF-β and PD-L1, in advanced squamous cell carcinoma of the head and neck (SCCHN). - ASCO Publications, accessed August 11, 2025, https://ascopubs.org/doi/10.1200/JCO.2021.39.15_suppl.6020
26. Phase I Trial of First-line Bintrafusp Alfa in Patients with Locally Advanced or Persistent/Recurrent/Metastatic Cervical Cancer - AACR Journals, accessed August 11, 2025, https://aacrjournals.org/clincancerres/article/30/5/975/734298/Phase-I-Trial-of-First-line-Bintrafusp-Alfa-in
27. Study of the Efficacy and Safety of the Bintrafusp Alfa in Previously Treated Advanced Malignant Pleural Mesothelioma (BIMES) - ClinicalTrials.gov, accessed August 11, 2025, https://www.clinicaltrials.gov/study/NCT05005429
28. Bintrafusp alfa Recruiting Phase 2 Trials for Metastatic Soft Tissue Sarcoma / Advanced Soft Tissue Sarcoma Treatment | DrugBank Online, accessed August 11, 2025, https://go.drugbank.com/drugs/DB15387/clinical_trials?conditions=DBCOND0060596%2CDBCOND0055864&phase=2&purpose=treatment&status=recruiting
29. Bintrafusp alfa Completed Phase 1 Trials for Small Cell Carcinoma / Small Cell Lung Cancer (SCLC) Treatment - DrugBank, accessed August 11, 2025, https://go.drugbank.com/drugs/DB15387/clinical_trials?conditions=DBCOND0036950%2CDBCOND0048944&phase=1&purpose=treatment&status=completed
30. Bintrafusp Alfa Program Rollover Study - ClinicalTrials.gov, accessed August 11, 2025, https://clinicaltrials.gov/study/NCT05061823
31. M7824 Versus Pembrolizumab as a First-line (1L) Treatment in Participants With Programmed Death-ligand 1 (PD-L1) Expressing Advanced Non-small Cell Lung Cancer (NSCLC) | ClinicalTrials.gov, accessed August 11, 2025, https://www.clinicaltrials.gov/study/NCT03631706
32. M7824 Versus Pembrolizumab as a First-line (1L) Treatment in Participants With Programmed Death-ligand 1 (PD-L1) Expressing Advanced Non-small Cell Lung Cancer (NSCLC) - larvol clin, accessed August 11, 2025, https://clin.larvol.com/trial-detail/NCT03631706
33. Clinical Trial: NCT03631706 - My Cancer Genome, accessed August 11, 2025, https://www.mycancergenome.org/content/clinical_trials/NCT03631706/
34. GSK, German Merck's $4.2B bintrafusp alfa drug flops again, but ..., accessed August 11, 2025, https://www.fiercebiotech.com/biotech/gsk-german-merck-s-4-2b-bintrafusp-alfa-drug-flops-again-but-companies-squint-to-see
35. Bintrafusp Alfa Shows Similar Efficacy to Pembrolizumab in PD-L1-High/Advanced NSCLC, accessed August 11, 2025, https://www.targetedonc.com/view/bintrafusp-alfa-shows-similar-efficacy-to-pembrolizumab-in-pd-l1-high-advanced-nsclc
36. Merck KGaA, Darmstadt, Germany, Reports Topline Data for Bintrafusp Alfa as Second-Line Monotherapy Treatment in Biliary Tract Cancer - PR Newswire, accessed August 11, 2025, https://www.prnewswire.com/news-releases/merck-kgaa-darmstadt-germany-reports-topline-data-for-bintrafusp-alfa-as-second-line-monotherapy-treatment-in-biliary-tract-cancer-301247680.html
37. GSK, Merck KGaA's BTC Drug, Bintrafusp Alfa Fails for Second Time This Year - BioSpace, accessed August 11, 2025, https://www.biospace.com/gsk-merck-kgaa-btc-drug-fails-for-second-time-this-year
38. Phase 2 trial of bintrafusp alfa as second-line therapy for patients ..., accessed August 11, 2025, https://pmc.ncbi.nlm.nih.gov/articles/PMC10442127/
39. Bintrafusp alfa, a bifunctional fusion protein targeting TGF-β and PD-L1, in advanced squamous cell carcinoma of the head and neck: results from a phase I cohort - The Journal for ImmunoTherapy of Cancer, accessed August 11, 2025, https://jitc.bmj.com/content/8/2/e000664
40. Bintrafusp alfa, a bifunctional fusion protein targeting TGF-β and PD-L1, in advanced squamous cell carcinoma of the head and neck: results from a phase I cohort - PMC, accessed August 11, 2025, https://pmc.ncbi.nlm.nih.gov/articles/PMC7342865/
41. Evaluation of bintrafusp alfa, a bifunctional fusion protein targeting TGF-β and PD-L1, in cervical cancer: Data from phase 1 and phase 2 studies. - ResearchGate, accessed August 11, 2025, https://www.researchgate.net/publication/352078222_Evaluation_of_bintrafusp_alfa_a_bifunctional_fusion_protein_targeting_TGF-b_and_PD-L1_in_cervical_cancer_Data_from_phase_1_and_phase_2_studies
42. Long-Term Therapy With Bintrafusp Alfa in HPV-Associated Cancers - The ASCO Post, accessed August 11, 2025, https://ascopost.com/news/september-2021/long-term-therapy-with-bintrafusp-alfa-in-hpv-associated-cancers/
43. Bintrafusp alfa Completed Phase 1 Trials for Solid Tumors / Metastatic Cancer Treatment, accessed August 11, 2025, https://go.drugbank.com/drugs/DB15387/clinical_trials?conditions=DBCOND0028461%2CDBCOND0029860&phase=1&purpose=treatment&status=completed
44. Metastatic Breast Carcinoma Completed Phase 1 Trials ... - DrugBank, accessed August 11, 2025, https://go.drugbank.com/indications/DBCOND0036429/clinical_trials/DB15387?phase=1&status=completed
45. Efficacy, safety, and biomarker analyses of bintrafusp alfa, a bifunctional fusion protein targeting TGF-β and PD-L1, in patients with advanced non-small cell lung cancer, accessed August 11, 2025, https://jitc.bmj.com/content/12/3/e008480
46. Selection of the Recommended Phase 2 Dose for Bintrafusp Alfa, a ..., accessed August 11, 2025, https://pmc.ncbi.nlm.nih.gov/articles/PMC7484916/
47. 465 Bintrafusp alfa in combination with chemotherapy in patients with stage IV NSCLC: safety and pharmacokinetic results of the INTR@PID LUNG 024 study | Journal for ImmunoTherapy of Cancer, accessed August 11, 2025, https://jitc.bmj.com/content/9/suppl_2/a494
48. Phase I Trial of First-line Bintrafusp Alfa in Patients with Locally Advanced or Persistent/Recurrent/Metastatic Cervical Cancer - Scientia, accessed August 11, 2025, https://scientiasalut.gencat.cat/bitstream/handle/11351/11170/phase_i_trial_first_line_bintrafusp_alfa_patients_locally_advanced_persistent_recurrent_metastatic_cervical_cancer_2024.pdf?sequence=1&isAllowed=y
49. Phase II evaluation of combination immunotherapy with CV301, N-803, bintrafusp alfa, and M9241 in patients with advanced small bowel and colorectal cancers. | Journal of Clinical Oncology - ASCO Publications, accessed August 11, 2025, https://ascopubs.org/doi/10.1200/JCO.2023.41.4_suppl.116
50. Phase I study of bintrafusp alfa, a bifunctional fusion protein targeting TGF-β and PD-L1, in patients with pretreated biliary tract cancer, accessed August 11, 2025, https://jitc.bmj.com/content/8/1/e000564
51. Bintrafusp Alfa: A Bifunctional Fusion Protein Targeting PD-L1 and TGF-β, in Patients with Pretreated Colorectal Cancer: Results from a Phase I Trial | The Oncologist | Oxford Academic, accessed August 11, 2025, https://academic.oup.com/oncolo/article/28/2/e124/6964580
52. Phase I study of bintrafusp alfa, a bifunctional fusion protein targeting TGF-β and PD-L1, in patients with pretreated biliary tract cancer - PubMed, accessed August 11, 2025, https://pubmed.ncbi.nlm.nih.gov/32461347/
53. Merck and GSK Announce Global Alliance to Jointly Develop and Commercialize M7824, a Novel Immunotherapy with Potential in Multiple Difficult-to-Treat Cancers, accessed August 11, 2025, https://www.merckgroup.com/en/news/m7824-2019-02-05.html
54. GSK and Merck Break Up $4 Billion Oncology Collab After Multiple Failures - BioSpace, accessed August 11, 2025, https://www.biospace.com/gsk-and-merck-kgaa-break-off-4-billion-oncology-collaboration-following-multiple-trial-failures
55. Phase 2 Study of Bintrafusp Alfa in Recurrent/Metastatic Olfactory Neuroblastoma (BARON)., accessed August 11, 2025, https://www.clinicaltrials.gov/study/NCT05012098
56. Despite the setbacks, interest in TGF-β remains | ApexOnco ..., accessed August 11, 2025, https://www.oncologypipeline.com/apexonco/despite-setbacks-interest-tgf-b-remains
57. Current landscape and future directions of bispecific antibodies in cancer immunotherapy, accessed August 11, 2025, https://www.frontiersin.org/journals/immunology/articles/10.3389/fimmu.2022.1035276/full
58. Application and future prospects of bispecific antibodies in the treatment of non-small cell lung cancer, accessed August 11, 2025, https://pmc.ncbi.nlm.nih.gov/articles/PMC12032835/
59. Application and future prospects of bispecific antibodies in the treatment of non-small cell lung cancer - ResearchGate, accessed August 11, 2025, https://www.researchgate.net/publication/390560154_Application_and_future_prospects_of_bispecific_antibodies_in_the_treatment_of_non-small_cell_lung_cancer