Why does treatment response vary?

Hypertrophic scars do not all respond to laser treatment in the same way: one may become softer and lighter, while another changes little—or some measures may even worsen. A research group in the United Kingdom compared response to an ablative fractional carbon dioxide (CO₂) laser with the cellular composition of scar tissue. The findings were published in Communications Medicine in 2026 [1].

Burns require medical care for approximately 11 million people each year [2], and scarring develops in 32–72% of those affected [3]. A fractional CO₂ laser creates microscopic zones of thermal injury separated by areas of undamaged skin. This triggers remodeling, with healing taking place from the preserved tissue.

How the study was conducted

The study included 14 adults with hypertrophic burn scars: 9 men and 5 women. Their mean age was 46.57 ± 4.23 years (range, 20–70), and their scars had been present for a mean of 21.71 ± 6.45 years (range, 2–65). The burned area averaged 50.31 ± 7.25% of total body surface area (TBSA), with a range of 10–85%.

The researchers used a Lumenis UltraPulse CO₂ system in SCAARFX and/or DeepFX modes. SCAARFX settings were 110–150 mJ at 3% density; DeepFX settings were 17.5–20.0 mJ at 5% density. In both modes, the shape was 2, the size was 10, and treatment was delivered as a single pulse at 300 Hz, with a 0.5-second interval and no active cooling. Each area was treated in one pass. The course consisted of three sessions at three-month intervals, with follow-up for up to one year after the first procedure.

This prospective pilot study followed participants according to a predetermined plan, and researchers assessed each patient’s scar before and after treatment. There was no separate untreated control group, and the study was exploratory.

To analyze the cells, the researchers used single-cell RNA sequencing (scRNA-seq) on 116,759 cells. The analysis included 16 scar biopsies from four participants—two with a good response and two with a weak response—as well as three samples of healthy skin. The 5-mm biopsies were collected at baseline, three weeks, six months, and one year.

Treatment response was assessed using the modified Vancouver Scar Scale (mVSS), ultrasound thickness measurements, a Cutometer, colorimetry, and the observer component of the Patient and Observer Scar Assessment Scale (POSAS-O).

Results

Differences in scar-cell composition accompanied differences in clinical response. The findings can be considered at both levels.

Cellular diversity in scar tissue

The analysis of cellular composition was based on biopsies from four participants—two with a good clinical response and two with a weak response. These exploratory findings come from a small sample, so the observed differences do not prove that cellular composition itself caused the treatment response [1].

• Scar tissue contains several cell types. In addition to fibroblasts, the researchers found vascular cells, pericytes, keratinocytes, melanocytes, Schwann cells, and lymphoid and myeloid cells. This shows that scar tissue is not a uniform mass of fibrous tissue but a complex cellular community.
• Fibroblasts are also diverse. The researchers identified five fibroblast subsets: secretory papillary, secretory reticular, pro-inflammatory, myofibroblasts, and mesenchymal fibroblasts. The proportions of these subsets differed among participants.
• In cases with a good clinical response, mesenchymal fibroblasts were more common, and genes associated with extracellular matrix organization—including COL14A1, POSTN, and SPARC—were more highly expressed. The authors link this profile to tissue remodeling and regenerative features.
• With a weak response, the proportion of pro-inflammatory fibroblasts was higher, and signs of inflammatory activity persisted. These included expression of CXCL14, JUN, and TNC, as well as genes in the HLA cluster in immune cells. Expression of the chemokines CXCL9 and CCL19 persisted for longer; inferred networks of interactions involving immune cells and endothelial cells also differed.

Overall, the findings point to different cellular profiles: matrix remodeling was more prominent with a good response, while persistent inflammation was more prominent with a weak response. Earlier publications have also discussed the role of chronic inflammation in pathological scarring [4, 5]; this study examined cellular features in the context of laser treatment.

How scar age related to laser treatment outcomes

We calculated changes as a percentage reduction in each measure: a positive value indicated improvement, while a negative value indicated worsening. The results for scars less than six years old and older scars, respectively, were:

• Skin thickness by ultrasound: 14.12 ± 7.08% vs. −23.00 ± 13.30%.
• Overall mVSS score: −0.48 ± 5.93% vs. −25.48 ± 8.99%.
• Vascularity: −1.88 ± 9.78% vs. −57.29 ± 19.24%.
• Overall observer score on POSAS-O: 15.71 ± 8.12% vs. −32.14 ± 12.91%.
• POSAS-O thickness score: 1.67 ± 13.94% vs. −59.52 ± 13.04%.

For all the between-group differences listed above, p < 0.05. Scar age correlated with the ranking of outcomes: r = 0.6974, p = 0.0071; older scars were ranked less favorably.

For older scars, the group averages shifted in the direction of worsening: thickness, vascularity, overall mVSS scores, and POSAS-O scores were, on average, worse than before treatment. But even among the younger scars, not all measures improved: mVSS and vascularity also worsened, though to a much lesser degree. The statistically significant findings were differences between the groups, not changes within each group. These are group averages, not predictions for individual patients.

Color changed in both groups, with no significant between-group differences. Across the full sample, color difference from baseline (ΔE) increased from 5.93 ± 0.76 at three months to 19.64 ± 2.70 at one year (p = 0.0005). However, no significant improvement across the full sample was found for most of the other measures.

What limits the conclusions

Most of the molecular findings came from RNA analysis. This shows which genes are active in cells, but it does not directly establish how much of the corresponding proteins is present. At the protein level, the authors tested only collagens I and III and tissue inhibitor of metalloproteinases 1 (TIMP1). In the weak-response group, TIMP1 gene activity was elevated, but it was not accompanied by a corresponding increase in protein levels. Changes in gene activity therefore cannot automatically be treated as confirmed changes in scar protein composition.

The study included only 14 participants, and single-cell RNA sequencing was performed on samples from four of them. This is not enough to establish the observed cellular features as reliable predictors of treatment response. The youngest scars in the study were two years old, so the results do not describe the response of recently formed scars.

There was no untreated control group. As a result, it is difficult to distinguish procedure-related changes from changes that might have occurred in the scar tissue independently of treatment. The authors did not report adverse-event data. This does not mean that no complications occurred; it means the study does not allow treatment safety to be assessed.

Practical takeaways

Taken together, the study findings suggest several factors associated with treatment response:

• Scar age was the only clinical factor studied that significantly associated with outcome. Consider it when planning a course of fractional CO₂ laser treatment and discussing expectations, without treating the six-year boundary as a universal patient-selection criterion.
• Fibroblast subsets: mesenchymal fibroblasts were associated with a good response, while pro-inflammatory fibroblasts were associated with a weak response.
• Inflammation and immune-cell composition: a weak treatment response was accompanied by persistent immune activity.
• Cell-to-cell signaling. With a weak response, RNA analysis suggested a possible increase in some signaling connections among fibroblasts, immune cells, and vascular endothelial cells. These connections may contribute to persistent inflammation and fibrosis, but their activity and impact on treatment outcomes remain unconfirmed.
• Gene-expression profile: COL3A1 and COL1A1 expression was higher with a good response, while inflammatory programs were more active with a weak response.

The study found no significant differences in response associated with patient sex, age, scar location, burn area, or body mass index. However, only 14 people participated, so the lack of statistically significant differences does not mean these factors do not affect treatment outcomes.

Differences in cellular composition and gene activity offer clues to why scars respond differently to laser treatment. The authors discuss whether weak responses might one day be improved by additionally targeting processes involved in inflammation and fibrosis. Potential targets include the myostatin (MSTN) signaling pathway, as well as TIMP1, CXCL14, and COL3A1, which are linked to matrix regulation and inflammatory processes. The study did not test these interventions, but the observations point to directions for further research.

References

1. Chen Y.-Y., Mahony C., Turner J., et al. Cellular heterogeneity in hypertrophic burn scars in response to carbon dioxide laser therapy. Commun Med (Lond) 2026; 6: 336.
2. World Health Organization. Burns. Fact sheet. 2018.
3. Lawrence J.W., Mason S.T., Schomer K., Klein M.B. Epidemiology and impact of scarring after burn injury: a systematic review of the literature. J Burn Care Res 2012; 33(1): 136–146.
4. Ogawa R. Keloid and hypertrophic scars are the result of chronic inflammation in the reticular dermis. Int J Mol Sci 2017; 18(3): 606.
5. Schaffrick L., Ding J., Kwan P., Tredget E.E. Molecular features of hypertrophic scars after thermal injury: is there a biologic basis for laser therapy? Adv Wound Care (New Rochelle) 2022; 11(4): 163–178.