FGF7: A Literature Course on What the Published Studies Report

This course organises the published literature on FGF7 into six modules. It describes what researchers studied, in which models, and what they reported — nothing more. This page is for educational purposes only and is not medical advice; consult a licensed physician before making any health decision. No protocol, quantity, schedule or personal application is suggested anywhere on this page, and each module closes with an explicit statement of where the evidence runs out.

Module 1 — What FGF7 Is and How It Has Been Studied

Definition and class

FGF7 is a member of the fibroblast growth factor (FGF) family of secreted signalling proteins. In the older literature and in product labelling it is more often called keratinocyte growth factor (KGF). It is a protein growth factor rather than a small synthetic peptide, and it is grouped with FGF3, FGF10 and FGF22 in what the structural literature calls the FGF7 subfamily, a set of ligands defined by their shared receptor preference and by structural features described in a 2019 review of the subfamily's structural biology (PMID 30809251).

Origin and where it is expressed

FGF7 is an endogenous mammalian protein; it is not a designer compound. It is generally described as a mesenchyme-derived ligand that acts on neighbouring epithelium. Recent work has extended that picture to other tissue compartments: a 2024 single-cell RNA sequencing study of skeletal muscle reported an interaction between muscle satellite cells and fibro-adipogenic progenitors mediated by FGF7 signalling (PMID 38751367), and a 2025 study reported autocrine FGF7/FGFR2 signalling within astrocytes in the central nervous system (PMID 41049737).

Forms that appear in the literature

• Native FGF7/KGF — the endogenous protein, usually studied through its receptor interactions and through expression mapping, as in the 2019 structural review (PMID 30809251).
• Recombinant human FGF7 — produced for laboratory use; a 2024 biotechnology report described construction and characterisation of a functional variant of human FGF7 generated by circular permutation and reported enhanced properties for the engineered version (PMID 38528341).
• Truncated or derivative molecules — a 2022 physiology report examined a molecule the authors designated FGF7p in the context of urothelial protection against cyclophosphamide (PMID 35748317).

How it has been studied

The methods in this literature are mostly preclinical and mechanistic: X-ray/structural analysis and receptor-binding characterisation in the 2019 review (PMID 30809251), protein engineering and in vitro characterisation in the 2024 variant study (PMID 38528341), single-cell transcriptomics in the 2024 muscle study (PMID 38751367), and animal injury or toxin models in the tendon (PMID 41455730), bladder (PMID 35748317) and MPTP neurodegeneration (PMID 41049737) reports.

Limits of the evidence for Module 1: the verified literature summarised here defines FGF7 biochemically and shows where it is expressed, but none of these six reports is a human clinical trial of an FGF7 product. Nomenclature is also inconsistent across decades of publications — KGF, KGF-1 and FGF7 may refer to the same gene product, while engineered variants and derivative molecules are distinct entities that should not be read as interchangeable with the native protein.

Module 2 — Mechanism as Described in the Literature

Receptor engagement

The organising mechanistic idea in this field is receptor selectivity. The 2019 structural biology review of the FGF7 subfamily described the structural determinants that govern how these ligands engage fibroblast growth factor receptors and their splice variants, and framed the subfamily's biology in terms of that binding specificity (PMID 30809251). Because FGF7 subfamily ligands are described as acting preferentially on the "b" splice isoform of FGFR2, the classical mechanistic model is paracrine: mesenchymal cells release the ligand, and epithelial cells expressing the receptor respond.

Downstream signalling

Two of the verified reports name intracellular pathways. The 2022 urothelial study examined the role of AKT in FGF7p-mediated protection of the urothelium against cyclophosphamide and framed AKT signalling as a mechanistic element of the protective effect it reported (PMID 35748317). The 2025 central nervous system study described astrocyte FGF7/FGFR2 autocrine signalling as the mechanism linking FGF7 to neuroinflammation in its model (PMID 41049737), which is notable because it places FGF7 in an autocrine rather than a purely paracrine arrangement.

Cell–cell communication in repair tissue

The 2024 single-cell study framed FGF7 as a communication signal between two resident cell populations in skeletal muscle, reporting a previously undescribed interaction between muscle satellite cells and fibro-adipogenic progenitors mediated with FGF7 signalling (PMID 38751367). In connective tissue, the 2025 tendon study reported that FGF7 promoted load-bearing tendon regeneration while suppressing fibrosis (PMID 41455730), positioning the ligand at the intersection of regenerative and fibrotic tissue responses.

Engineering the molecule

Mechanism can also be altered deliberately. The 2024 circular permutation study rearranged the primary sequence of human FGF7 to generate a functional variant and reported enhanced properties relative to the parent protein (PMID 38528341), an approach that reflects how protein-engineering groups attempt to address stability or activity limitations of native growth factors.

Limits of the evidence for Module 2: mechanistic pathways were described in cells and animal tissue, not in humans, and the pathways named in one tissue do not automatically operate in another. The direction of effect differed by context — regenerative in tendon (PMID 41455730) and degeneration-promoting in the astrocyte model (PMID 41049737) — so no single mechanistic summary covers the whole literature.

Module 3 — Reported Outcomes by Study

Reading the pattern

Across these reports, FGF7 was studied as a tissue-context-dependent signal rather than as a general-purpose agent. Researchers reported a pro-regenerative, anti-fibrotic profile in load-bearing tendon (PMID 41455730) and a protective profile in urothelium exposed to cyclophosphamide (PMID 35748317), while the same ligand family was reported to drive neuroinflammation and dopaminergic neuron loss in a toxin model of parkinsonism (PMID 41049737). The muscle sequencing work added a descriptive layer, mapping FGF7 signalling onto a specific cell–cell relationship rather than testing an intervention (PMID 38751367).

Limits of the evidence for Module 3: these are laboratory and animal endpoints — histology, sequencing readouts, tissue markers and cell counts — not patient-reported outcomes, function scores or survival in humans. None of the verified reports establishes a benefit for any person, and none of them was replicated across independent groups within this verified set. Effects described in a single model, at a single time point, in a single species are hypothesis-generating findings.

Module 4 — FGF7 Side Effects: What Studies Report

None of the six verified reports was designed primarily as a safety or toxicology study, so this module describes the harm-relevant findings that appear in the published biology rather than an adverse-event table from a clinical trial.

Harm-direction findings

The most direct harm-relevant result in this set is neurological. The 2025 study reported that astrocyte FGF7/FGFR2 autocrine signalling mediated neuroinflammation and promoted MPTP-induced degeneration of dopaminergic neurons (PMID 41049737), meaning FGF7 signalling was positioned as a contributor to injury in that model rather than a protective factor. That finding sits alongside reports of the opposite direction in other tissues, such as the protective urothelial result described in the 2022 cyclophosphamide study (PMID 35748317) and the anti-fibrotic tendon result reported in 2025 (PMID 41455730).

Why receptor biology matters to risk discussion

Because the 2019 structural review described the FGF7 subfamily in terms of its receptor-binding specificity and structural determinants (PMID 30809251), any growth-factor ligand of this class acts on receptor-bearing cell populations wherever those receptors are expressed — which is the general reason FGF-family agents are studied with attention to unintended proliferative or inflammatory signalling. Similarly, the 2024 engineered-variant report described altered protein properties for a circularly permuted hFGF7 (PMID 38528341); a modified molecule is a different entity with its own unstudied safety profile.

Limits of the evidence for Module 4: there is no adverse-event frequency, no dose–toxicity relationship and no human tolerability data in the verified set. Absence of reported harm in a mechanistic or sequencing study is not evidence of safety, because those studies did not look for harm. Readers evaluating the safety of any approved FGF7-based product should consult the current prescribing information for that product and a licensed clinician.

Module 5 — Pharmacokinetics Where Data Exist

Pharmacokinetics is the weakest area of this literature. None of the six verified reports published human absorption, distribution, metabolism or elimination parameters for FGF7, and no half-life, clearance or bioavailability figure can be drawn from them.

Two reports touch on time-related and stability-related questions without being formal pharmacokinetic studies. The 2022 physiology report examined the durability of FGF7p-mediated urothelial protection against cyclophosphamide, making duration of effect an explicit endpoint (PMID 35748317), which is a pharmacodynamic rather than a pharmacokinetic measure. The 2024 protein-engineering study reported that a circularly permuted variant of human FGF7 had enhanced properties compared with the unmodified protein (PMID 38528341), the kind of work that is typically motivated by handling and stability constraints of recombinant growth factors.

General protein pharmacology explains why oral figures are absent: FGF7 is a protein of substantial size, as reflected in the structural descriptions of the subfamily (PMID 30809251), and proteins of this class are not expected to survive gastrointestinal digestion intact. The verified literature does not test that question directly.

Limits of the evidence for Module 5: no pharmacokinetic parameter of any kind is reported in the verified set, and nothing on this page should be read as implying a duration of action, an exposure level or a route of administration for any person. Anyone seeking pharmacokinetic data for an approved recombinant KGF product would need to consult that product's regulatory documentation.

Module 6 — Regulatory Status

The following is general regulatory information, not legal advice.

Approved products

Recombinant human keratinocyte growth factor — the protein encoded by the FGF7 gene — has been developed into a prescription biologic marketed under the name Kepivance (palifermin), a truncated recombinant form authorised in the United States for reducing the incidence and duration of severe oral mucositis in patients with haematologic malignancies receiving myelotoxic therapy requiring haematopoietic stem cell support. Marketing status, labelling and availability differ by jurisdiction and change over time, so current official product information and national regulatory databases are the authoritative sources. None of the six verified reports summarised in this course is a registration trial for that product.

Research-use-only material

Recombinant FGF7 and engineered FGF7 variants are widely supplied as research reagents labelled research use only (RUO). RUO labelling signifies that the material has not been evaluated or authorised as a drug for human administration and is intended for laboratory investigation. The circularly permuted variant described in 2024 is an example of a laboratory construct rather than an authorised medicine (PMID 38528341), as is the FGF7p molecule used in the 2022 urothelial experiments (PMID 35748317).

Compounding

In the United States, pharmacy compounding operates under sections 503A and 503B of the Federal Food, Drug, and Cosmetic Act, which restrict which bulk drug substances may be compounded and under what conditions. Biologics licensed under the Public Health Service Act sit outside the ordinary compounding pathways available for small-molecule drug substances, and a research-grade growth factor is not, by virtue of being purchasable as a reagent, a compoundable drug substance. State boards of pharmacy add further requirements.

Limits of the evidence for Module 6: regulatory status is a moving target and is not evidence of efficacy or safety for any indication beyond an approved label. The existence of an approved KGF-based biologic for one narrow oncology-supportive-care indication says nothing about the tendon, muscle, bladder or neurological findings described in Modules 2 and 3.

What the Studies Did Not Test

• Healthy human volunteers. None of the verified reports administered FGF7 to healthy people or measured outcomes in them.
• Any dose, route or schedule for human use. The verified set supports no quantity, frequency or duration for a person, and none is stated on this page.
• Cross-tissue generalisation. The tendon regeneration finding (PMID 41455730) was not tested against the neurodegeneration finding (PMID 41049737) in a single organism to determine net effect.
• Long-term exposure and cancer surveillance. No verified report followed animals or people long enough to characterise chronic proliferative risk.
• Combinations. No verified report examined FGF7 alongside other peptides, growth factors or medications outside the specific cyclophosphamide challenge context (PMID 35748317).
• Comparative effectiveness. FGF7 was not compared with standard care for tendon injury, sarcopenia, cystitis or Parkinson's disease in the verified set.
• Special populations. Pregnancy, paediatric, renal or hepatic impairment scenarios were not addressed.

The honest summary of this literature is that FGF7 is a well-characterised endogenous growth factor with an active preclinical research programme in several unrelated tissues, one approved narrow-indication biologic derived from it, and no verified human evidence base for the regenerative or protective uses that its animal findings suggest. This page remains educational only and is not medical advice; a licensed physician is the appropriate source for individual decisions.

References

• Structural Biology of the FGF7 Subfamily (Frontiers in Genetics, 2019)
• Durability of and role of AKT in FGF7p urothelial protection against cyclophosphamide (Physiological Reports, 2022)
• Construction and characterization of a functional variant hFGF7 with enhanced properties by circular permutation (Biotechnology Journal, 2024)
• Single-cell RNA-seq reveals novel interaction between muscle satellite cells and fibro-adipogenic progenitors mediated with FGF7 signalling (Journal of Cachexia, Sarcopenia and Muscle, 2024)
• Astrocyte FGF7/FGFR2 autocrine signaling mediates neuroinflammation and promotes MPTP-induced degeneration of dopaminergic neurons (Acta Pharmaceutica Sinica B, 2025)
• FGF7 promotes load-bearing tendon regeneration and suppresses fibrosis (Nature Communications, 2025)