Ki-67 is a protein associated with cell proliferation which increases during the mitotic process, mainly in the S phase of the cycle. The expression of Ki-67 in tumor cells, quantified in percentage, is called the Ki-67 index, which could be associated with prognosis in tumors such as melanomas, and breast and bladder carcinomas.12-15
In 2000, Shimizu, et al16 showed a relevant correlation between the Ki-67 index and choline levels on MR spectroscopy images. However, their research had limitations, such as a small number of patients, image artifacts and errors in spectral evaluation.16,17
Our study revealed a positive association between the Ki-67 index and edema zone volume in GBM. To the best of our knowledge, this is the first study to find this statistical correlation. Volume of necrosis zone and total tumor volume showed no association with a high proliferation index as shown by other authors.17-19
The method for volumetric calculations in this study was quantitative, semi-automatic, which offers the advantage of manually demarcating the contours of each type of volume selected in each of the regions of interest (ROI) of an image. This allowed us to avoid deviations in lesion volume that commonly occur when a completely automatic methodology is applied.19,20
According to Odland et al20 the disadvantage of the semi-automatic method is an interobserver variation when quantifying volume. Thus, in order to avoid divergences in volume quantification, two neurosurgeons performed the volumetric analysis in the present work.
We chose Horos Project because it is an open source, easy-to-handle software and it is accurate to delimit tumor compartments.21
In 2019, Henker, et al17 tried to show associations between tumor compartment volume and Ki-67 index in a study with 150 glioblastoma patients, using a 3D neuronavigation software. The authors, however, did not succeed, possibly due to a variation of volume measurement methodologies and the software used. Nevertheless, this same study found associations between elevated Ki-67 index and glioblastoma outcome.
Armocida, et al22 presented an association between high Ki-67 index and large total volumes (>45 cm3) but not edema volumes, unlike our study. In this case, despite the use of similar measurement methods and software, compartment definition, which is not clearly stated, could differ from ours.
Another divergence between findings of different studies is the proliferative index variation according to the site of extraction for biopsy, as proposed by Jakovlevs, et al.23
An interesting finding from Armocida, et al22 was a negative association between Ki-67 index and patient progression-free survival.
Other authors support the idea of an association between the Ki-67 index and prognosis. Liang, et al24 showed a correlation between the Ki-67 index and first-year mortality in 335 glioma patients and Chen, et al11 revealed in their meta-analysis that elevated levels of Ki-67 can be a predictive factor for poor prognosis in glioblastoma.
Therefore, we could propose a negative association between edema volume and prognosis in patients with glioblastoma. Based on this, patients with large volumes of edema should undergo not only urgent surgery, but oncological evaluation during hospitalization and adjuvant treatment as soon as possible. This could contribute to better outcomes.
Despite these findings, a limitation of our research is the retrospective format and, therefore, prospective studies are necessary to clarify the relation between glioblastoma compartment volume and clinical variables such as prognosis.