Reply to Markaryan et al. Comment on “Aji et al. aMMP-8 POCT vs. Other Potential Biomarkers in Chair-Side Diagnostics and Treatment Monitoring of Severe Periodontitis. Int. J. Mol. Sci. 2024, 25, 9421”

1. Introduction

We appreciate Markaryan et al. [1] for raising timely and important queries on the role of aMMP-8 (active-matrix metalloproteinase-8) and tMMP-8 (total latent 75 and 55 kDa proMMP-8) in periodontal diagnostics and our ability to compare them. MMP-8 in periodontitis- and peri-implantitis-affected gingiva, gingival crevicular fluid (GCF) and peri-implant sulcular fluid (PISF), mouth rinse, and saliva is converted to its 10 kDa lower-molecular-size active forms and, at the same time, further fragmented (Figure 1A–C), whereas the MMP-8 in the oral fluids of healthy gingival tissues, gingivitis, and mucositis exists mostly as the latent or total MMP-8 pro-form. The same MMP-8 activation of proMMP-8 has been demonstrated in wound healing in patients who have undergone, for example, excimer laser photorefractive keratectomy [2,3]. aMMP-8 is a catalytically competent collagenolytic and proteolytic immunomodulator [4,5,6]. In contrast, the latent, total, or pro 75 and 55 kDa MMP-8 forms are catalytically inactive, lacking collagenolytic and proteolytic activity. This fundamental distinction has been rigorously demonstrated through various collagen degradation diagnostic assays conducted before the advent of the antibody-based enzyme-linked immunosorbent assay (ELISA), the time-resolved immunofluorescence assay (IFMA), and chair-side immunoassay techniques [7,8,9,10].

The role of aMMP-8 as a key biomarker in periodontitis and its diagnostic potential have been extensively illustrated in previous research [11,12,13,14,15]. Studies using collagen degradation activity assays have provided strong evidence supporting this conjecture [7,8,9,10]. These foundational studies form the scientific basis for subsequent research and development and support the necessity of anti-MMP-8 polyclonal and monoclonal antibodies, as well as the development of both immunoassays and chair-side assays. Key studies that have contributed to this field include those by Sorsa et al. [16] and Mäntylä et al. [17], both of which utilized and successfully validated a diagnostic cut-off value of 20 ng/mL for aMMP-8 detection. Penttala et al. [18] recently had success in utilizing an aMMP-8 mouth rinse cut-off value of 20 ng/mL in an algorithmic mobile diagnostic procedure capable of delivering results in 5 min and 1s. AI applications [18] would not have been possible with cut-offs at 10 ng/mL [19] and 25 ng/mL [20,21].

Nonetheless, variations in cut-off values have been reported in some studies. For instance, Deng et al. [19] have explored lower thresholds such as 10 ng/mL, while Lorenz et al. [20] have used 25 ng/mL. Despite these variations, independent studies conducted in India, Sweden, Greece, Finland, Germany, Holland, and Nigeria, by researchers including Aji et al. [22], have consistently validated the 20 ng/mL cut-off as an effective and optimal threshold for periodontitis risk assessment. Notably, Aji et al. [23] demonstrated that cut-offs of 10 ng/mL and 25 ng/mL do not provide reliable diagnostic accuracy. These findings underscore the robustness of the 20 ng/mL cut-off, reinforcing its clinical reliability and widespread adoption in non-invasive periodontal screening methodologies.

2. Comments and Discussion

In response to the first point by Markaryan et al. [1], the monoclonal antibody utilized in the aMMP-8 point-of-care test (POCT) PerioSafe/Oralyzer and IFMA is the same as the one described in Sorsa et al.’s US patents. This antibody is designed to detect active MMP-8, its activation products, and related lower-molecular-weight fragments at the same time (Figure 1A–C). Independent studies have also confirmed these findings using both monoclonal and polyclonal anti-MMP-8 antibodies [9,17]. While Lee et al. [7] did not investigate antibodies but mainly collagen degradation assay in their research, Romanelli et al. [9] provided further validation of these results through independent monoclonal antibody studies. Additionally, activation-related lower-molecular-size MMP-8 fragments (Figure 1A–C) are believed to not be collagenolytic, as shown by Hasty et al. [24] and Knäuper et al. [25], but formed from aMMP-8 during activation. Periodontitis- or kidney disease-related activation with concomitant further fragmentation of active MMPs (Figure 1A–C) is not only a characteristic of MMP-8 but also concerns MMP-9, -13, and -14 [9,24,25].

In response to the second, third, and fourth points, our findings indicate that tMMP-8 ELISAs are not effective in diagnosing periodontitis or serving as a reliable biomarker within the new 2017 classification system [26,27,28,29], which is related to a significant extent to the work by Sorsa et al. in terms of utilizing aMMP-8 as a biomarker for this system for periodontal or peri-implant disease [22]. Instead, clinicians and researchers should prioritize methods that specifically target aMMP-8, as it provides a more precise reflection of ongoing collagen breakdown, inflammation, and clinically progressive soft and hard tissue destruction associated with periodontal and peri-implant diseases [7,8,9,10,15,23,30,31,32,33,34,35].

We compared the aMMP-8 POCT Oralyzer chair-side immunotest with the catalytic active collagenase activity assay [22]. The Academic Editor’s comment on this comparison is both relevant and valuable. In response, we emphasize the following: even if the MMP-8 antibody can detect both active and inactive MMP-8, the detection method (FRET substrate) enables only the measurement of the active form of MMP-8 [22]. The FRET substrate was not claimed to be specific to MMP-8. The specificity of the assay comes from the use of the capture antibody [22]. Our objective was to compare POCT with an entirely independent catalytic activity assay, which would not have been feasible if the same antibody had been used in both methods [22].

Furthermore, multiple international and independent cohort studies have consistently shown that aMMP-8 POCT and IFMA outperform total MMP-8 ELISA in diagnostic precision. Across all assays, aMMP-8 POCT and IFMA have demonstrated superior diagnostic accuracy in relation to tMMP-8 ELISAs [16,22]. These studies, conducted across diverse international populations or cohorts and clinical settings, reinforce the clinical reliability and accuracy of aMMP-8-based assays in detecting active periodontal peri-implant destruction.

In the present study [22], full-mouth clinical parameters and biomarkers, i.e., the aMMP-8 POCT, the aMMP-8 RFU activity [22], total MMP-8 ELISA, myeloperoxidase, PMN elastase, TIMP-1, calprotectin, and interleukin-6, were analyzed and compared at baseline and after non-surgical therapy at 6 weeks (Figure 2). The aMMP-8 POCT with a cut-off at 20 ng/mL, but not with cut-offs of 10 ng/mL and 25 ng/mL, and myeloperoxidase were the most efficient in discriminating between periodontal health and disease (measured by ROC AUC [22]) followed closely by aMMP-8 RFU activity and PMN elastase. In comparison, total MMP-8 R&D ELISA was found a little less efficient in that regard, while TIMP-1, calprotectin, and interleukin-6 were the least precise. Similar findings were found in the treatment effect monitoring ([23] and Figure 2). Additionally, GCF samples from six healthy and six periodontitis-affected sites (Stage III/IV, Grade B/C) were analyzed via Western blotting for MMP-8 using a polyclonal anti-MMP-8 and monoclonal anti-MMP-8 antibodies (Figure 1A–C). Western blotting results clearly show that in healthy GCF, MMP-8 is predominantly in its latent (total) pro-form. However, in periodontitis-affected GCF, MMP-8 is consistently and without exception converted to its active form, with the concomitant formation of lower-molecular-weight fragments, as confirmed by both polyclonal and monoclonal antibodies (Figure 1A–C). Finally, there is the dento-ELISA with Sorsa et al.’s US patent-19 anti-aMMP-8-monoclonal antibodies, as well [16]. Thus, targeting these MMP-8 activation species can be achieved conveniently by multiple diagnostic methods utilizing Sorsa et al.’s US patent-19 antibodies in oral fluids and human excretion fluids [3,16].

3. Conclusions

These findings clearly indicate that active MMP-8, together with its lower-molecular-weight fragments, serves as a biomarker of MMP-8 activation in periodontitis (Figure 1 and Figure 2), further supported by the independent international studies [23]. Previous studies have shown that collagenolytic aMMP-8 with a cut-off at 20 ng/mL—but not the non-collagenolytic total MMP-8—can be implemented into the 2017 periodontitis classification system as a biomarker of the staging and grading of periodontitis (Figure 3) [23,29]. Thus, we should not synonymize aMMP-8 and total MMP-8 in periodontitis diagnostics. The aMMP-8-test technology is commercially available as IFMA, dento-ELISA, RFU, and POCT-Oralyzer chair-side assays [16,22,23].