155 MET, ECD shedding, and loss of E-cadherin positivity, it remains higher than the threshold set for WTSs (results not shown). This is also the case for ECD shedding. Similar to the lack of association between uniform negativity and survival for WTSs, we argue that these differences in thresholds – for both D1C2 uniform positivity and ECD shedding – are likely due to TMA sampling and tumor heterogeneity (17, 18). Finally – in contrast to the TMA results – the WTS study shows that shedding is not only associated with DFS but also associated with OS. Because shedding is determined within D1C2 uniform positive tumor patches for the WTS study, this finding is in line with the TMA result, showing that uniform positivity is associated with OS and DFS (17, 18). Despite these differences, the study shows that other findings are consistent with prior results. The uncorrected HRs for both D1C2 uniform positivity and ECD shedding are of the same order of magnitude for the WTSs and TMA studies (Supplementary tables 22 and 23). In addition, the HR found for OS for the uniform negative NCH-38 staining pattern is comparable with the HR reported for loss of E-cadherin described in a meta-analysis concerning E-cadherin immunoreactivity in OSCC (29). It is therefore concluded that the developed scoring system provides valid results. Established use of scoring systems (30) and/or CDx (31, 32) using patterns and intensity scoring in the field of pathology indicates that it is feasible to implement the developed scoring system in a – oral cancer – diagnostic setting. The observation that C-terminal MET uniform positivity and ECD shedding are independently associated with poor OS and DFS – independent of disease stage (Supplementary tables 24 and 25) in TM C-terminal MET-positive OSCC – concurs with the fact that ECD shedding has been described to increase the malignant potential of the MET oncogene (33). Moreover, it suggests that MET is a promising target for therapy. However, low success rates of performed clinical trials led to the belief that immunohistochemistry is inadequate for patient stratification. Instead, it is argued that stratification should be based on MET genetic aberrations, such as amplification, point mutations, exon 14 skipping, and oncogenic fusions (34, 35). Indeed, recent trials implementing patient selection based on genetic alterations show initial successes in tumor reduction. However, inhibition of wildtype MET activity has been shown to reduce cell survival, local invasion, and distant metastasis. This makes wildtype MET a suitable target for adjuvant therapy after curative primary surgery as its targeting potentially eradicates residual cancer cells (34). Taking everything into consideration and knowing that reliable antibodies were used, we think that the results presented here could be of added value in the development of CDx (Figure 4). 5
RkJQdWJsaXNoZXIy MTk4NDMw