Evaluation of the novel monoclonal antibody against DOG1 as a diagnostic marker for gastrointestinal stromal tumors

Abstract EN

Background and Objectives : Gastrointestinal stromal tumors (GISTs) are the most common mesenchyme tumors of the gastrointestinal tract.

With the growing effectiveness and availability of first and second-generation tyrosine kinase inhibitor (TKI) drugs, the accurate diagnosis of GIST has become imperative.

The problem is that some GISTs with KIT or Alpha-type platelet-derived growth factor receptor (PDGFRA) mutations may have low KIT expression by immunohistochemistry yet will still benefit from TKI drugs.

Molecular analysis is a costly and laborious process.

Therefore the emergence of a new sensitive immunohistochemical marker for GISTs would be ideal.

Recently antibodies against "Discovered on GIST-1" (DOG1) have been generated.

The aim of this study was to evaluate the monoclonal DOG1.1 antibody as a diagnostic marker for GISTs and to compare immunohistochemical staining and diagnostic efficacy of DOG1.1 with that of KIT in GISTs.

Materials and Methods : Forty seven paraffin embedded GISTs were immunostained with both KIT and DOG1.1 antibodies.

Immunoreactivity was graded semi quantitatively from 0 to 4.

Some other mesenchyme tumors were included in the study and stained for both markers to test for their specificity Results : out of the 47 GISTs, 44 were immunoreactive for both KIT and DOG1.1 antibodies (93.62 %).

Two cases (4.25 %) were KIT-positive DOG-negative and the remaining case was DOG-positive KIT-negative (2.13 %).

A statistically significant concordance was found between KIT and DOG1.1 immunoreactivity (p = 0.004), with moderate agreement between immunostaining scores (kappa = 0.379).

As regards tumor site, a statistically significant association was found between high DOG1.1 scores and gastric GIST (p = 0.008).

High KIT and DOG1.1 immunostaining scores were significantly associated with high-risk tumors (p = 0.002 and p = 0.002 respectively).

DOG1.1 immunoreactivity was focal in more than half of the cases.

The overall diagnostic accuracy of DOG1.1 was 96.5 %, with a specificity and sensitivity of 100% and 95.7 % respectively.

The overall positive predictive values and negative predictive values were 100 % and 84.6 %, respectively.

As for KIT, The overall diagnostic accuracy was 94.8 %, with a specificity and sensitivity of 81.8% and 97.8 %, respectively.

The overall positive predictive values and negative predictive values were 95.8 % and 90 %, respectively.

Conclusions : this study has demonstrated that DOG1.1 is a more specific marker for GIST than KIT, yet its sensitivity is a little inferior to that of KIT owing possibly to the focal staining of DOG1 antibody in many cases.

In conclusion, in our institution, in which cost effective health care is a priority, we recommend using DOG1 as the first choice antibody for the diagnosis of GIST as it is a more specific marker.

If it is negative, then KIT is tried.

If this latter is also negative, then several other antibodies should be tested.

And owing to the potential therapeutic significance of GIST diagnosis, if still the results of immunohistochemistry are ambiguous, mutational analysis should be considered to confirm the diagnosis.