A scar that won't back down

Keloids hold a unique position among connective tissue pathologies. Unlike hypertrophic scars, which may regress over time, keloids not only persist but continue to expand beyond the boundaries of the original wound — often months or years after healing. They are frequently accompanied by itching, burning, pain, and significant psychological distress. While the prevalence of pathological scarring in the general population is estimated at 1–2%, it is substantially higher among individuals with darker skin phenotypes.

Intralesional injections remain the first line of treatment. Triamcinolone acetonide (TAC), a corticosteroid, is injected directly into the scar tissue either alone or in combination with other agents. 5-fluorouracil (5-FU), a cytotoxic agent that inhibits fibroblast proliferation and collagen synthesis, serves as a common alternative. Both approaches are supported by clinical trials and included in current guidelines. However, experienced practitioners know that response remains unpredictable. Some scars flatten and soften after just a few sessions; others, seemingly identical, do not react at all. The question "why?" has long lacked a definitive answer.

One hypothesis links this variation to the immune microenvironment. It is well established that inflammation and dysregulated immune responses play a central role in keloid pathogenesis [2]. Macrophages, T-lymphocytes, and regulatory T-cells (Tregs) are all present in the scar tissue, yet their specific contribution to a keloid's "fate" under different treatments has remained insufficiently understood.

Study objectives and design

A Finnish research group from the University of Tampere set out to determine whether a keloid’s immune profile dictates its response to intralesional therapy and how this profile shifts depending on whether treatment is successful.

The study was based on a completed double-masked randomized controlled trial (RCT) involving 48 patients with established keloids who received either TAC (20 mg/ml mixed 1:1 with 10 mg/ml lidocaine) or 5-FU (50 mg/ml). Injections were administered at 3–4 week intervals for a maximum of three procedures per course, continuing until a response was achieved or the limit was reached [3]. Biopsies were taken from the active peripheral margin of the scar — the point of most aggressive growth — both before and after the treatment course.

Biopsy samples were analyzed using immunohistochemistry and quantified via digital virtual microscopy. Researchers evaluated the density and distribution across dermal layers (superficial, middle, and deep) of several cell populations: M1 and M2 macrophages (CD163 marker for M2), CD4+ and CD8+ T-lymphocytes, regulatory T-cells (Tregs; FOXP3 marker), neutrophils, dendritic cells, Schwann cells, and high endothelial venules.

A treatment response was recorded in 25 out of 48 patients: 15 in the TAC group and 10 in the 5-FU group. Consequently, the non-responder group consisted of 10 (TAC) and 13 (5-FU) patients.

Key findings: an unexpected start and a telling conclusion

The initial results were surprising: none of the cell populations studied in the pre-treatment biopsies correlated with the subsequent response to therapy. In other words, it is currently impossible to "read" an untreated scar’s immune profile to predict whether it will respond to injections. This finding is significant as it closes the door — at least at our current level of knowledge — on using pre-therapeutic immune profiles as predictive biomarkers.

However, the post-treatment landscape told a completely different and more informative story. In responders, there was a significant increase in anti-inflammatory M2-polarized macrophages (CD163+). Across the entire biopsy, the concentration in responders was 228 cells/mm² compared to 119 cells/mm² in non-responders (p=0.045). In the middle dermis, this difference was even more striking: 97 vs. 37 cells/mm², respectively (p=0.017). The most pronounced accumulation of M2 macrophages occurred in the TAC group, suggesting that corticosteroid therapy most actively reshapes the immune microenvironment toward an anti-inflammatory phenotype.

Furthermore, a statistically significant inverse correlation was found between M2 macrophage count and myofibroblast density in the middle dermis (r=−0.51; p=0.0008). Myofibroblasts are the primary effector cells in fibrosis, synthesizing the excessive collagen that defines the density and volume of scar tissue [4]. Previous research indicated that the middle dermis contains the highest concentration of keloid myofibroblasts [3]. This inverse relationship is a critical discovery, suggesting that the recruitment of anti-inflammatory macrophages is linked to the suppression of fibrotic remodeling.

Another notable finding involved regulatory T-cells (Tregs). In the superficial dermis of responders — particularly in the TAC group — the number of Tregs after treatment was significantly lower than in non-responders (p=0.027). The role of Tregs in keloids is complex; while they suppress immune responses, their accumulation also correlates with increased fibrosis [5]. Their reduction in the context of a successful response may reflect a normalization of the local immune balance, though this mechanism requires further study.

Other cell populations, such as neutrophils, dendritic cells, and Schwann cells, showed no significant changes in any group.

Limitations

The authors acknowledge several limitations. The sample size is relatively small (48 patients, 25 responders).

Additionally, the study only covers two types of intralesional therapy, meaning the results cannot be extrapolated to other methods like laser therapy or surgical excision.

Finally, the specific biological mechanisms driving the observed M2 polarization and Treg reduction were not explored and remain a subject for future research.

Practical takeaways

For clinicians managing scars, these data are vital on several levels:

1. They provide a biological explanation for a well-known clinical phenomenon: two identical-looking keloids can respond differently to the same treatment due to their internal immunological context.
2. The link between M2 macrophages and myofibroblast regression highlights a potential new direction for therapy. If M2 accumulation is a driver of fibrosis resolution rather than just a byproduct, stimulating this polarization could complement standard protocols.
3. The stronger immune shift in the TAC group indirectly supports the use of corticosteroids when a significant restructuring of the scar’s inflammatory environment is the priority.

Conclusion

This study demonstrates that keloid response to intralesional therapy is associated with a profound reshaping of the immune microenvironment. Anti-inflammatory M2 macrophages accumulate in the dermis of responding patients and inversely correlate with the density of myofibroblasts, the key cells of fibrosis.

While baseline immune profiles do not yet allow us to predict treatment success, these data provide the first "internal" look at a successful therapeutic response and identify the immune mechanisms accompanying it. This work points toward a promising direction for future research into more reliable keloid treatments. The next step will be investigating the functional states of CD163+ macrophages and the mechanisms behind their recruitment in responding scars. Only then can we determine if targeted modulation of the macrophage response could form the basis of new therapeutic strategies for keloid management.

References

1. Komulainen T., Ylitörmä M., Hietanen K.E. et al. Anti-inflammatory macrophages are recruited to keloids that respond to intralesional injection therapies. J Invest Dermatol 2026;146(5):1438–1442.
2. Lee C.C., Tsai C.H., Chen C.H. et al. An updated review of the immunological mechanisms of keloid scars. Front Immunol 2023; 14: 1117630.
3. Komulainen T., Daymond P., Hietanen K.E. et al. Myofibroblasts reside in the middle dermis of the keloids but do not predict the response to injection therapies: a double-blinded, randomized, controlled trial. Front Med 2024; 11: 1293028.
4. Kokubo K., Onodera A., Kiuchi M. et al. Conventional and pathogenic Th2 cells in inflammation, tissue repair, and fibrosis. Front Immunol 2022; 13: 945063.
5. Chen Y., Jin Q,. Fu X. et al. Connection between T regulatory cell enrichment and collagen deposition in keloid. Exp Cell Res. 2019; 383(2): 111549.