The Use of KI-67mmunohistochemistry in the Classification and Grading of Central Nervous System (CNS) Tumors in Tanzania

Abstract

Background: The 2016 World Health Organization classification for central nervous system tumors includes assessment of molecular markers along with histology. As a consequence, some new diagnostic entities have been added, but on the other hand, variants and patterns that have no diagnostic and/or biological relevance are no longer recognized. However, this classification cannot absolutely be implemented without modification in majority of low-and middle-income countries such as Tanzania because of unavailability of molecular testing such as isocitrate dehydrogenase wild and mutant types which are among of essential parameters for the classification. Therefore, in resource constrained countries like Tanzania, an additional simple and affordable method of grading the central nervous system tumors is inevitable to compliment the current classification for prognostication purposes and for planning of appropriate patient’s management. Conversely, the immunoreactivity of Ki-67 protein has been associated with the proliferation of cell populations in various tumors permitting it to be used as a marker of tumor aggressiveness. This study aims at demonstrating the applicability of Ki-67 labelling index as an added tool to support the 2016 World Health Organisation classification of Central Nervous System tumors in centers where facilities for molecular testing are lacking. Objective: To evaluate the use of Ki-67 immunohistochemistry in the classification and histological grading of central nervous system tumors. Materials and methods: This was a retrospective cross-sectional study where all central nervous system tumors diagnosed at Muhimbili National Hospital between June 2009 and June 2019 were studied. All cases with retrievable patient information as well as anatomical location were included. Each tumor was reviewed histologically and classified according to 2016 WHO grading system. Ki67 immunohistochemistry was done by utilizing tissue micro-array method where the most hyper-cellular area with minimal or no necrosis was sampled for tissue microarray processing. Ki-67 labelling index was analyzed by choosing an area with high proliferation index as assessed by Ki-67 positive tumor cells with nuclear stain counted in 100 tumor cells in a continuous manner. The number of positive cells was considered to be the Ki-67 labelling index of the case, where a 20% grade IV respectively. Results: A total of 141central nervous system tumor cases were enrolled in this study. There were 77 (54.6%) males and 64 (45.6%) females with male to female ratio of 1.2:1. The age ranged from 1 to 77 years with overall mean age of 37.6 + 18.8 years, and median of 40 years. Children below 10 years old contributed to 11.3% of all cases. The frontal lobe was mostly affected consisting of 26/141 (18.4%) cases, followed by temporal lobe 19/141 (13.5%) and dura-based lesions 16/141 (11.3%). About 55.3% (78/141) of all tumors were meningiomas of which grade I comprised of 85.9% (67/78), followed by astrocytic tumors 22.7% (32/141) and ependymal tumors 4.9% (7/141). About 64.5% tumors were grade I histologically, 22.7% grade II, 0.7% grade III and 12.1% grade IV, however Ki-67 LI index showed grade I tumors to be 52.4% with a KI-67 LI range of 0 – 8%, grade II, 31.4% with Ki-67 LI ranging from 3% – 14%, grade III composed of 6.4% with Ki-67 LI range from 7% - 28% and grade IV 9.3% with Ki-67 LI of 14% - 90%. There was about 22.2% of grade I tumors with higher proliferation index, furthermore about 71.4% of all primary tumors which initially was diagnosed to be grade I histologically had a Ki-67 LI of 5-9% whereas 85.7% of their counterpart recurrent tumors had a Ki-67 LI of 5-9% and they were all grade I histologically. Conclusion: Ki-67 labelling index has the potential of detecting grade I tumors with aggressive behaviors and therefore subjecting these patients to a more appropriate treatment hence significantly reducing the risks of recurrence and improving the overall patient survival. Furthermore, the study showed that in Tanzania the most common central nervous system tumors are meningiomas followed by astrocytomas. The most common affected region was supratentorial with frontal lobe leading as the most affected anatomical site. This is to our knowledge the first study in Tanzania to analyze all central nervous system tumor types involving patients of all age groups consequently forming a baseline ‘data and a reference for further epidemiological and clinico-pathological studies in the country and the region at large.