The Regional Expression Of HIF-1α In Retinoblastoma And Its Relationships With Proliferation,Apoptosis And Angiogenesis

Background:Retinoblastoma (Rb) is the most common intraocular malignancy in children. With the application of chemoredution and focal adjuvant therapies (transpupillary thermotherapy, TTT; cryotherapy; subconjunctival carboplatin, SCC; Plaque radiotherapy; etc), it is possible to salvage the eyes and optimize the residual vision in less advanced tumors. However, there still are cases with recurrence or metastasis after combined therapy. On the one hand, a crucial factor for the phenomenon is the chemotherapy resistance. Studies have shown that anoxia in retinoblastoma may related to the drug resistance, for the reason that tumor cells in the anoxic areas are insensitive to the traditional chemotherapy drugs. On the other hand, according to the respective features of the different therapies, the therapeutic modes and effects may vary in different tumor zones in retinoblastoma, which may produce treatment blind spots and eventually results in recurrences and drug resistances. Thus, studies of anoxia in different tumor zones in retinoblastoma may be of great importance not only in searching the mechanism of drug resistance, but in revealing the reasons for the after-combined-therapy recurrence, in order for us to find a optimized therapeutic mode to increase the curative effects and improve prognosis.Part OneThe regional expression of HIF-1αin retinoblastoma and its relationships with proliferation, apoptosis and angiogenesisObjective To detect the regional expression of hypoxia inducible factor-1α(HIF-1 a) in retinoblastoma and the its relationships with VEGF, Bax and Ki-67 and try to search the reasons for recurrences and drug resistances.Methods Immunohistochemical study for HIF-1α, VEGF, Bax and Ki-67 was performed in 39 paraffinaceous examples of retinoblastoma. Every pathological section was divided into five regions:the surface region, the central part, the bottom part, the choroidal region and seeding tumors. The expressions, correlations and distributional differences of these factors were all invested both integrally and regionally.Results In the 39 cases of retinoblastoma,29 cases were positive for HIF-1α(74.4%), particularly as a whole,10 cases (25.6%) for (-),17 cases (43.6%) for (+),12 cases (30.8%) for (++); regionally, HIF-1αwas positive in 71.1%,36.8%,84.2%, 54.5% and 82.1% of the cases in the surface region, the central part, the bottom part, the choroidal region and seeding tumors, respectively, which was statistically different (P<0.001). The positive rate of VEGF, Bax and Ki-67 was 53.8%,66.7% and 59.0%, respectively. In different regions, the positive rates of VEGF and Bax were different (both P<0.001), but there was no regional distinctions in the expression of Ki-67 (P=0.976). As a whole, both the expression of VEGF and Bax had a positive correlation with that of HIF-1α(rs=0.48, P=0.002 and rs=0.39, P=0.021), but there was no relationship between the expressions of Ki-67 and HIF-1α(P=0.606). Regionally, the expressions of VEGF, Bax and HIF-1αshared similar distributional features:positive rates were higher in the surface region, bottom part and seeding tumors, and lower in the central part and choroidal region, which was different form the expression of Ki-67.Conclusions The anoxic zones are more likely to be located in the marginal parts in retinoblastoma, which may due to fewer blood vessels in these parts than in the central ones during tumor formation. In addition, hypoxia may increase the level of apoptosis and promote the formation of new blood vessels, so as to endure the anoxic environment, and succeed in avoiding the effects of the traditional chemotherapy agents. Studies on hypoxia may be of great importance in understanding and solving the drug resistance in retinoblastoma.Part TwoBlade-scraping TestObjective To eliminate the influences of the edge effect and man-made factors during experiments.Methods Based on the former immunohistochemical study for HIF-1α,5 paraffinaceous examples which are more likely to be stained in the surface region than in the central part were chosen. The tumor edge of one slide of every example was scraped by a blade for the blade-scraping group; another slide was not scraped for the untreated group; still another was used as a control. Repeat the immunohistochemical staining of HIF-1αunder the same conditions.Results After blade-scraping, the new surface regions in the treated group were not stained as dark as that in the untreated group, while the staining of the central parts in both groups were roughly the same.Conclusions Based on the Blade-scraping test, avoidance of dry slides, random diagnosis, etc., the influences of the edge effect and man-made factors during experiments can be avoided. Thus, the outcome of the study is reliable:the distribution features of anoxia in retinoblastoma reflect its own characteristics.