BLMP6: UTHealth's Fibulin-4-Targeting Peptide for Metastatic Breast Cancer

What Was Published

A team at UTHealth Houston has reported that a small peptide called BLMP6 can find metastatic breast cancer cells in the body, can carry an imaging dye to them for detection, and can deliver a cytotoxic drug to them in mice with measurable survival benefit. The paper appeared in Molecular Therapy Oncology in late April 2026. Its formal title is "Fibulin-4 expressed in metastatic breast cancer is a target of peptide-based imaging probes and experimental therapeutics." First author is Alexes C. Daquinag; the senior author is Mikhail Kolonin, who directs the Center for Metabolic and Degenerative Diseases at McGovern Medical School. The paper carries the DOI 10.1016/j.omton.2026.201207.

The molecule of interest is a synthetic peptide named BLMP6 that binds a protein called fibulin-4. Fibulin-4 is a component of the extracellular matrix, abundant in connective tissue, and — the new finding — expressed at high levels on aggressive invasive breast cancers but not on normal breast tissue or non-invasive tumors. That difference is what makes it useful as a target. The headline result is two-part: BLMP6 conjugated to a fluorescent dye localized to metastatic lesions in mice grafted with triple-negative breast cancer; BLMP6 conjugated to monomethyl auristatin E (MMAE), an FDA-approved cytotoxic warhead, suppressed metastasis and improved survival.

The work was led by Mikhail Kolonin's lab at UTHealth Houston in collaboration with the McGovern Medical School and the McKelvey School. A UTHealth news release and Medical Xpress write-up went out April 29, 2026.

Key Facts

• Paper: Daquinag AC, Kolonin M, et al. Molecular Therapy Oncology, 2026. DOI 10.1016/j.omton.2026.201207.
• Lead institution: The Brown Foundation Institute of Molecular Medicine, McGovern Medical School at UTHealth Houston.
• Senior author: Mikhail Kolonin, PhD, Harry E. Bovay Jr. Distinguished University Chair in Metabolic Disease Research.
• The peptide: BLMP6, a synthetic short peptide identified by in vivo phage display, binds the extracellular matrix protein fibulin-4.
• The target: Fibulin-4 (FBLN4), highly expressed in aggressive invasive breast cancers, absent or low on normal breast tissue and non-invasive lesions.
• Imaging result: BLMP6 labeled with a fluorescent dye traveled directly to metastatic lesions in mice xenografted with triple-negative breast cancer cells.
• Therapeutic result: BLMP6 conjugated to monomethyl auristatin E (MMAE), an FDA-approved cytotoxic, suppressed metastasis and improved survival in xenograft mice.
• Translation evidence: Tissue arrays from human breast cancer patients confirmed BLMP6 binding correlated with invasive disease but not with non-invasive cancer or healthy tissue.
• AI involvement: The team used computational structural modeling, including AlphaFold-derived structures, to predict and verify the BLMP6-fibulin-4 interaction.
• Stage: Preclinical. No human trials initiated. No IND filing reported.

Why This Paper Matters

Most breast cancer deaths are caused by metastasis rather than the primary tumor. The primary mass can be surgically removed and treated with adjuvant therapy. The cells that have already escaped to distant sites — bone, liver, lung, brain — are harder to find, harder to image, and harder to drug. Antibody-drug conjugates have improved survival in HER2-positive metastatic disease, and recent ADCs like trastuzumab deruxtecan have changed the standard of care. Triple-negative breast cancer, which lacks the receptors that ADCs typically target, has not benefited as much. Several recent approvals have moved the field forward, but the patient population that needs new options is still large.

What the Kolonin team has done is take a peptide-driven approach to the same problem. A peptide that binds fibulin-4 with reasonable selectivity gives you two capabilities at once. With a fluorescent payload, the peptide becomes an imaging probe that lights up metastatic deposits. With a cytotoxic payload, the same peptide becomes a delivery vehicle for chemotherapy that concentrates the drug at the lesions and reduces exposure elsewhere in the body. Both applications are practical. Both have working preclinical data in this paper.

The choice of fibulin-4 as a target is the part that makes the work biologically interesting. Fibulin-4 is not a tumor antigen in the classical sense. It is a structural protein, part of the elastic fiber assembly machinery, present in connective tissue throughout the body. The team's finding is that fibulin-4 expression is reorganized around aggressive invasive tumors in a way that is not seen in normal breast tissue or in non-invasive tumors. The peptide does not seek out "breast cancer" in the abstract. It seeks out a feature of the tumor microenvironment that correlates with invasive behavior.

How BLMP6 Was Found

The Kolonin lab has spent two decades using in vivo phage display to find peptides that home to specific tissues and disease states. The method is direct. A phage library encoding around 10^9 distinct peptide sequences is injected into a mouse model. After some time, the tissue of interest is harvested, the phage particles that have lodged there are recovered, and their peptide-coding sequences are read out. After several rounds of enrichment, the peptides that bind preferentially to the target tissue dominate the pool. The peptides are then synthesized and tested.

For BLMP6, the target tissue was metastatic breast cancer in xenograft mice. After enrichment, the team identified a short peptide sequence that bound preferentially to the disseminated tumor cells. Standard biochemistry then identified the binding partner: fibulin-4. The team used a combination of pull-down assays, surface plasmon resonance, and AI-driven structural modeling to characterize the interaction. The AI part is worth flagging. Computational structural prediction has become standard for understanding small-peptide-to-protein interactions; the team used predicted structures of fibulin-4 to identify the binding interface and confirmed the prediction with mutagenesis.

The translational step is what separates this from a typical academic paper. The team did not stop at "BLMP6 binds fibulin-4 in mice." They obtained human breast cancer tissue arrays representing the full range of disease — non-invasive, early invasive, late invasive, metastatic — and stained them with labeled BLMP6. The binding pattern in human tissue matched the binding pattern in mice. The correlation between BLMP6 binding and aggressive disease was clear; the correlation with non-invasive or normal tissue was minimal.

The Drug Conjugate

Monomethyl auristatin E is a synthetic analog of dolastatin-10, a microtubule inhibitor too toxic to administer systemically as a free drug. MMAE only became clinically useful when it was put on the end of an antibody, where the antibody handles targeting and the MMAE handles killing. Brentuximab vedotin and several other approved ADCs use this approach. The first MMAE-based ADC reached the market in 2011; the class is now mature.

What the Kolonin team has done is replace the antibody with a peptide. The advantages are well-known and well-rehearsed in the peptide-drug-conjugate literature: peptides are smaller, penetrate tissue more deeply, are easier and cheaper to manufacture, and can be designed with defined chemistry. The disadvantages are equally well-known: peptides have shorter serum half-lives than antibodies and can be cleared rapidly by the kidney. Whether a peptide-drug conjugate is a workable format for a given indication depends on whether the half-life problem can be solved with linkers, PEGylation, or repeated dosing.

In the BLMP6-MMAE results, mice with triple-negative breast cancer xenografts that received the conjugate showed suppressed metastasis to distant organs and improved survival compared with controls. The paper reports the data in detail; the headline is consistent with what a working PDC ought to do. The team did not formally compare BLMP6-MMAE against standard-of-care chemotherapy or against approved ADCs. That comparison will be necessary before any clinical decision can be made about whether the molecule offers a real advantage.

Why Fibulin-4 Is an Interesting Target

Most targeted cancer therapies hit antigens that distinguish tumor cells from normal cells — HER2 on certain breast cancers, CD19 on B cells, PSMA on prostate. Fibulin-4 is not that kind of target. It is a normal protein, expressed throughout the body in connective tissue. What makes it useful here is not its presence on tumor cells but its altered expression pattern around invasive tumor margins.

The biology is consistent with what cancer cell biologists have understood for the better part of a decade. Aggressive tumors restructure their extracellular matrix. They cause stromal cells to produce different mixes of collagen and elastin. They alter fibulin family expression. Fibulin-4 specifically gets upregulated. The result is a target signature that lives in the tumor microenvironment rather than on the tumor cell itself. BLMP6 binds this microenvironment-defined signal.

The clinical implication is interesting. If fibulin-4 is upregulated by the process of invasion itself, then a BLMP6-based imaging probe could in principle detect not just established metastases but very small invasive lesions that have started restructuring their environment but have not yet reached imaging thresholds for current standard-of-care techniques. That is speculative based on what the paper reports. It is the kind of speculation that motivates the next round of work.

What This Paper Doesn't Show

The work is preclinical. All efficacy data are from mouse xenograft models. Xenografts under-represent the heterogeneity of human disease and the immune component is removed in the immunocompromised mice typically used for these experiments. Human triple-negative breast cancer behaves differently than its mouse-implanted version in several ways that affect drug delivery.

The team did not report dose-finding or formal toxicology beyond what was needed for the proof-of-concept experiments. They did not report pharmacokinetics in any detail. Peptide-drug conjugates often have short serum half-lives that require frequent dosing or linker-based solutions; whether BLMP6-MMAE has a workable PK profile for clinical use is unknown from this paper.

Fibulin-4 expression in non-cancer disease is real. Some inherited connective tissue disorders involve fibulin-4 dysregulation, and remodeling fibrosis in non-cancer contexts could in principle produce off-target BLMP6 binding. The paper's tissue array data are reassuring on the cancer side but do not exhaustively characterize what BLMP6 looks like in patients with concurrent connective tissue disease.

Finally, no IND filing has been reported. The team's stated next step is clinical translation, which typically means animal toxicology in a second species, scaled manufacturing, formulation, and the IND-enabling package. None of that is publicly underway at the time of publication.

What This Means for Different Audiences

For oncologists treating metastatic triple-negative breast cancer: the molecule is years away from any patient-level decision. Current standard of care — chemotherapy, immune checkpoint inhibitors in PD-L1-positive disease, sacituzumab govitecan — remains unchanged. The new work does not affect treatment selection today.

For the broader peptide-drug-conjugate field: the paper is a useful proof point. PDCs have been a quieter category than ADCs but the format has clear advantages in tissue penetration and manufacturing cost. A working preclinical example against a major indication, with an FDA-approved cytotoxic warhead and a defined molecular target, gives the field a concrete reference.

For patients following metastatic breast cancer research: this is one of perhaps a dozen interesting preclinical programs at any given time. Most preclinical results, however promising, do not reach clinic. The realistic interpretation is that this work justifies further investment in BLMP6 and the fibulin-4 target, not that a new drug is imminent.

For peptide chemists: the in vivo phage display + structural prediction + tissue array validation pipeline is increasingly the template for peptide therapeutic discovery against complex targets. The Kolonin lab's combination of physical screening with computational structure prediction is a worth-studying example.

Frequently Asked Questions

Is BLMP6 a treatment for breast cancer?

Not yet. It is a research peptide that has shown effects in mice. It is not in clinical trials, has no IND on file, and is not available outside the lab. The paper shows that the conjugate suppresses metastasis and improves survival in a mouse model of triple-negative breast cancer. Whether it can do anything similar in humans is unknown.

How does BLMP6 differ from antibody-drug conjugates like trastuzumab deruxtecan?

Antibody-drug conjugates use a full antibody to target the cytotoxic payload. They are large, expensive to manufacture, and slower to penetrate solid tumors. BLMP6 is a small synthetic peptide with the same role as the antibody. Smaller molecules can penetrate tumor tissue more efficiently and are cheaper to make. The trade-off is that peptides typically have shorter serum half-lives. Both formats have a place; the BLMP6 work is evidence that the peptide-conjugate format can work for fibulin-4-positive disease.

What is fibulin-4 and why does it matter?

Fibulin-4 is a normal extracellular matrix protein that helps assemble elastic fibers in connective tissue. The team found that fibulin-4 is reorganized and upregulated in the tissue around invasive breast tumors. The peptide binds this rearranged fibulin-4 selectively. The target is therefore not the cancer cell itself but a feature of the tumor microenvironment that correlates with invasive behavior.

What is monomethyl auristatin E and why use it?

MMAE is a synthetic compound that disrupts microtubule assembly and kills dividing cells. As a free drug it is too toxic for systemic use. When attached to a targeting molecule — an antibody or, here, a peptide — it becomes selective enough to be useful. It is the cytotoxic warhead in several FDA-approved antibody-drug conjugates, which is what makes it a reasonable choice for an experimental peptide-drug conjugate.

How was BLMP6 discovered?

The Kolonin lab used in vivo phage display, a method where a library of around a billion peptide sequences is injected into a mouse model and the peptides that bind the tissue of interest are recovered and sequenced. After multiple rounds of enrichment, BLMP6 emerged as the peptide that bound preferentially to metastatic breast cancer cells. Computational structural prediction and biochemistry then identified fibulin-4 as the binding partner.

What is the next step?

Animal toxicology in a second species, formal pharmacokinetic studies, formulation work, and the IND-enabling package. The team has stated an intent to translate the work clinically. The realistic timeline for a first-in-human trial from this stage is two to three years; the work is at a typical academic preclinical milestone.

Related Coverage on PeptideKnow

• Another recent example of academic peptide discovery against a complex target: Suga Lab's RaPID-ExCells and HL4 (May 13, 2026)
• An academic peptide therapeutic moving toward the clinic: Entera Bio EB613 (May 11, 2026)
• Full profile of the lead molecule from this paper: BLMP6 fibulin-4-binding peptide
• The broader research-peptide landscape: Healing and recovery peptides

Sources

• Daquinag AC, Zhang S, An Z, Azhdarinia A, Ghosh S, Farmer S, Ramesh AK, AghaAmiri S, Hernandez Vargas S, Kolonin M. "Fibulin-4 expressed in metastatic breast cancer is a target of peptide-based imaging probes and experimental therapeutics." Molecular Therapy Oncology. 2026. DOI 10.1016/j.omton.2026.201207.
• Medical Xpress coverage: "Preclinical efficacy of experimental peptide therapy suggests a new target for metastatic breast cancer treatments," April 29, 2026.
• Kolonin Lab at UTHealth Houston: Mikhail Kolonin, PhD, McGovern Medical School.
• Background on in vivo phage display for peptide discovery: Kolonin MG, et al. "Reversal of obesity by targeted ablation of adipose tissue." Nature Medicine. 2004. PubMed PMID 15170203.
• Background on MMAE-based antibody-drug conjugates: Senter PD, Sievers EL. "The discovery and development of brentuximab vedotin for use in relapsed Hodgkin lymphoma and systemic anaplastic large cell lymphoma." Nature Biotechnology. 2012. PubMed PMID 22781697.
• Fibulin-4 in tumor extracellular matrix: Hsiao CT, et al. "Fibulin-4 in tumor progression." Frontiers in Oncology. 2023. PubMed PMID 37346073.

This article reports on preclinical research published in a peer-reviewed journal. BLMP6 is a research peptide. It is not approved for human use and is not available outside research settings. Nothing here is medical advice.