(032) Update on malignant transformation rate of oral epithelial dysplasia: a systematic review.

Background: Oral epithelial dysplasia (OED) is a term defining architectural changes in epithelial cells, knowingly associated with an increased risk of malignant transformation. The malignant transformation rate (MTR) has been reported in numerous previous studies, including meta-analyses; however, discrepancies in eligibility criteria among these studies have been observed. A critical review is essential to elucidate the current literature and establish the MTR of OED incorporating specific details.

Objectives: To update the latest information on MTR from OED.

Methods: Systematic review was performed with meta-analysis. PubMed, EMBASE, Cochrane databases were searched; we considered papers published until December 2023. Studies that examined the MTR of OED were eligible. Study selection and quality assessment were performed by two reviewers. Histologic grades of dysplasia, and disease modality (oral lichen planus [OLP], oral leukoplakia [OLK]) were factors considered to affect MTR. MTR was calculated as the incidence of malignant cases over the number of OED cases. Analysis was conducted based on Freeman-Tukey transformed proportion with random effects model. STATA SE 18.0 was used for all the analyses.

Results: Fifty-four studies were included in analysis. The pooled overall MTR was 16.0% (95% confidence interval: 12.0%, 20.3%). Heterogeneity between studies was significant (I^2=96.21%).
Twenty-nine studies reported tertiary histologic grade of OED. The pooled MTR of severe OED was 23.6% (11.4%, 38.0%), was 12.2% (6.2%, 19.6%) for moderate OED, and was 6.3% (3.9%, 9.3%) for mild OED. The observed positive linear trend for the histologic grade of OED was statistically significant (p < 0.01).
Eight studies reported binary histologic grade of OED. The pooled MTR of high grade OED was 41.3% (27.7%, 55.6%), was 13.5% (9.9%, 17.6%) for low grade OED.
Twenty-five studies reported OED with OLP&OLK. OLP was reported only in 2 studies. Twenty-three studies concerning OLK yielded MTR of 16.9% (12.8%, 21.4%, I^2= 83.36%). The MTR by disease modality was not different from the overall pooled MTR.

Conclusions: Our result demonstrated higher estimate of MTR than the previously reported 10.5% or 12.1% from recent meta-analysis. The differences may stem from the broader inclusion criteria applied in our analysis, which also facilitated the exploration of a subgroup analysis based on disease modality. The MTR of each disease modality did not show a significant difference from the overall MTR. We confirmed a statistical significance of linear trend for the histologic grade of OED.
Since there was large significant heterogeneity remaining between studies, additional analysis concerning contributing factors to MTR should be needed.