Phosphohistone H3 Immunohistochemical Labeling: a Potentially Useful Tool for Risk Stratication and Prognostic Analysis in Gastrointestinal Stromal Tumors.


 BackgroundIt is well recognized that risk stratification of gastrointestinal stromal tumors (GISTs) is closely related to tumor size, mitotic index (MI), and primary location. Among these three parameters, tumor size and primary location are easily established, while MI is subjective and its repeatability is poor. It is thus necessary to identify a biomarker to represent the true MI. Expression status and biological or prognostic significance of mitotic marker phosphohistone H3 (PHH3) and cell proliferation marker Ki67 in GIST have not been clearly understood until now. MethodsAn immunohistochemistry experiment was performed to detect the expression status of PHH3 and Ki67 in 125 paraffin-embedded GIST samples. All of the patients were followed up until September 30, 2019. ResultsThe MI determined using stained hematoxylin and eosin (H&E) sections (MI-H&E) and immunohistochemically positive PHH3 index (PHH3-IHC) was compared among groups of different genders, ages, primary locations, and histological subtypes, showing that the difference was not statistically significant (P>0.05). MI-H&E and the immunohistochemically positive Ki67 index were positively correlated (r=0.273, P=0.001), but the correlation was lower than that with the PHH3-positive index (r=0.705, P=0.000). The PHH3-positive index was also positively correlated with the Ki67 index (r=0.224, P=0.006). MI-H&E were positively correlated with MI quantified using immunohistochemically stained PHH3 sections (MI-PHH3) (P<0.05). After using PHH3 to perform MI quantification, the risk stratification of five GIST cases was changed, where two cases were given a higher risk grade and three cases were given a lower risk grade. Follow-up data were obtained from 98 cases (98/125, 78.4%), including two cases of metastasis and one death. Both metastatic and death cases had high MI-H&E. One metastatic case and one death case had high PHH3-positive indexes, while the remaining metastatic case had a low PHH3-positive index. ConclusionImmunohistochemical PHH3 labeling is a potentially useful tool for risk stratification and prognostic analysis in GIST. Using immunohistochemical PHH3 labeling makes it more convenient for pathologists to determine the MI for GIST. MI quantification with immunohistochemical PHH3 sections can be used as an adjunct tool for risk stratification and prognostic analysis of GIST, but cannot completely replace MI quantification using stained H&E sections. The Ki67 index is positively correlated with MI-H&E, although the efficiency of tumor risk stratification is lower than that of PHH3.


Abstract Background
It is well recognized that risk strati cation of gastrointestinal stromal tumors (GISTs) is closely related to tumor size, mitotic index (MI), and primary location. Among these three parameters, tumor size and primary location are easily established, while MI is subjective and its repeatability is poor. It is thus necessary to identify a biomarker to represent the true MI. Expression status and biological or prognostic signi cance of mitotic marker phosphohistone H3 (PHH3) and cell proliferation marker Ki67 in GIST have not been clearly understood until now.

Methods
An immunohistochemistry experiment was performed to detect the expression status of PHH3 and Ki67 in 125 para n-embedded GIST samples. All of the patients were followed up until September 30, 2019.

Results
The MI determined using stained hematoxylin and eosin (H&E) sections (MI-H&E) and immunohistochemically positive PHH3 index (PHH3-IHC) was compared among groups of different genders, ages, primary locations, and histological subtypes, showing that the difference was not statistically signi cant (P>0.05). MI-H&E and the immunohistochemically positive Ki67 index were positively correlated (r=0.273, P=0.001), but the correlation was lower than that with the PHH3-positive index (r=0.705, P=0.000). The PHH3-positive index was also positively correlated with the Ki67 index (r=0.224, P=0.006). MI-H&E were positively correlated with MI quanti ed using immunohistochemically stained PHH3 sections (MI-PHH3) (P<0.05). After using PHH3 to perform MI quanti cation, the risk strati cation of ve GIST cases was changed, where two cases were given a higher risk grade and three cases were given a lower risk grade. Follow-up data were obtained from 98 cases (98/125, 78.4%), including two cases of metastasis and one death. Both metastatic and death cases had high MI-H&E. One metastatic case and one death case had high PHH3-positive indexes, while the remaining metastatic case had a low PHH3-positive index.

Conclusion
Immunohistochemical PHH3 labeling is a potentially useful tool for risk strati cation and prognostic analysis in GIST. Using immunohistochemical PHH3 labeling makes it more convenient for pathologists to determine the MI for GIST. MI quanti cation with immunohistochemical PHH3 sections can be used as an adjunct tool for risk strati cation and prognostic analysis of GIST, but cannot completely replace MI quanti cation using stained H&E sections. The Ki67 index is positively correlated with MI-H&E, although the e ciency of tumor risk strati cation is lower than that of PHH3.

Introduction
Gastrointestinal stromal tumors (GISTs) are the most common non-epithelial tumors of the gastrointestinal tract. Compared to the epithelial tumors, the incidence of GISTs is lower. GIST presentation is insidious at the onset of the disease. Some patients have almost no clinical symptoms and the tumor is often detected during a routine physical examination or in association with other diseases. The incidence of GIST ranges from 6.8 to 20 per million people in different countries [1]. GIST can range from benign to malignant according to its biological behavior. The National Institutes of Health (NIH) classi es tumor risk strati cation of GIST into four levels: very low risk, low risk, moderate risk, and high risk. At present, according to the modi ed GIST classi cation standard of NIH 2008, GIST risk strati cation is closely correlated with tumor size, mitotic index (MI), and primary location.
Among the factors affecting GIST grading, MI quanti cation often differs among observers. Although there are some markers re ecting the proliferation index, none of them can correctly represent MI. Phosphohistone H3 (PHH3) is a newly studied eukaryotic core histone. Histone H3 begins to be phosphorylated in the mitotic G2 phase and dephosphorylated at the end of the mitotic (M) phase. Therefore, PHH3 is expressed in mitotic G2 and M phases [2,3]. Since Hendzel et al. [2] rst proposed using PHH3 as a mitotic marker in 1997, research on PHH3 in different tumors has been continuously reported.
However, the expression status and signi cance of PHH3 in GIST have rarely been reported until now. The present study used immunohistochemical (IHC) staining to explore the expression and signi cance of the mitotic marker PHH3 and cell proliferation marker Ki67 in GIST. The relationship among the expression status of PHH3 and Ki67, clinicopathological characteristics, tumor risk strati cation, and prognosis in GIST was analyzed.

Page 4/18
The present study was performed with Institutional Review Board's approval and all patients provided informed consent. A total of 125 consecutive, formalin-xed, and para n-embedded GIST samples with different grading were obtained from patients who had undergone an operation between January 2012 and December 2017 in the First People's Hospital of Zigong and between January 2016 and December 2018 in the Fourth People's Hospital of Zigong. None of the patients were administered treatment before the operation. Pathological diagnosis of all cases was con rmed by histomorphology and IHC staining. The haematoxylin-eosin (H&E)-stained sections were reviewed and the MI was independently re-calculated by two experienced pathologists before the experiment.

IHC
IHC staining was performed using an automatic IHC staining apparatus (Bench Mark GX, Roche, USA). Both polyclonal PHH3 antibody and monoclonal Ki67 antibody (clone number MIB-1) were purchased from Maixin Biotechnology Company (Fuzhou, China). Both antibodies were ready-to-use. Samples without primary antibodies were used as a blank control. Human lung adenocarcinoma tissue sections were used as a positive control.

Quantitative analysis of MI, PHH3-positive index, and Ki67 index
Evaluation of MI using H&E-stained sections (MI-H&E), PHH3-positive index, and Ki67 was performed sequentially in the same section areas. These parameters were observed using a light microscope (Olympus BX43, eld number: 22 mm). Two experienced pathologists chose a "hot spot" region to continuously observe 21 high-power elds ( 5 mm 2 ) and counted mitotic gures in the same area on H&Estained and PHH3 IHC-stained sections. The Ki67-positive index was calculated as the percentage of positive cells in a continuous counting of 1000 cells.

Result interpretation
The mean values for MI-H&E and PHH3-IHC evaluated by two pathologists were recorded. PHH3-positive cells taking up ≤ 5/5 mm 2 on immunohistochemically stained sections had a low expression, while those taking up > 5/5 mm 2 had a high expression. The percentage of Ki67-positive cells of ≤ 5% was recorded as low expression, while that of > 5% was recorded as high expression. On H&E-stained sections, the mitotic gure of ≤ 5/5 mm 2 was regarded as low MI (LMI), while > 5/5 mm 2 was regarded as high MI (HMI).

GIST risk strati cation
The GIST risk strati cation was performed according to the 2008 NIH modi ed GIST classi cation standard.

Follow-up
The follow-up was conducted on all included patients via telephone with a deadline of September 30, 2019. SPSS 23.0 (SPSS, Chicago, IL, USA) statistical software was used to perform the χ2 test on the classi cation data. One-way analysis of variance (ANOVA) was used for the comparison between groups.

Statistical analysis
Spearman's correlation test was used for correlation analysis between different groups. A P-value of < 0.05 was considered to indicate statistical signi cance.

Relationship between PHH3-positive index and clinicopathological characteristics of GIST
Immunolocalization of PHH3 was observed in the nuclei of GIST cells (Fig. 1). There were 16 GIST cases with a high PHH3-positive index (16/125, 12.8%). The PHH3-positive index was compared among groups of different genders, ages, locations, and histological subtypes, but the difference was not statistically signi cant (P > 0.05; Table 2).   (Table 3). Ki67 was positive in the nuclei of GIST cells (Fig. 2), and 65 GIST cases had a high Ki67 index (65/125, 52%). A total of 16 cases with HMI had a high Ki67 index, and 57 cases with LMI had a low Ki67 index. Spearman's correlation analysis results revealed that the Ki67 index and MI-H&E were positively correlated (r = 0.273, P = 0.001) ( Table 4), but the correlation was lower than that with the PHH3-positive index.

Correlation between PHH3-positive index and Ki67 index
In the 125 GISTs, 13 cases with a high PHH3-positive index had a high Ki67 index, and 57 cases with a low PHH3-positive index had a low Ki67 index. Spearman's correlation analysis results revealed that the PHH3positive index was positively correlated with the Ki67 index (r = 0.224, P = 0.006; Table 5). According to the Spearman's correlation analysis result, patient prognosis was positively correlated with both MI-H&E and PHH3-IHC (r = 0.506, P = 0.000; r = 0.314, P = 0.001, respectively; Table 6). In addition, patient prognosis was also positively correlated with the Ki67 index (r = 0.171, P = 0.047; Table 7), although this correlation was lower compared to both MI-H&E and PHH3-IHC.

MI quanti cation comparison between two groups
In the present study, two experienced pathologists determined the MI using double-blind reviewing of the enrolled cases. MI-H&E and PHH3-IHC results determined by the two pathologists were evaluated using univariate analysis. There was no statistical difference (P > 0.05).
3.8 Difference in GIST risk grading depends on MI-H&E and PHH3-IHC MI quanti cation and risk grading were performed on both H&E-stained sections and PHH3 IHC-stained sections, respectively. In 125 GIST cases, the risk strati cation of ve cases was changed and two cases were given a higher risk grade. These cases went from a very low-risk to moderate-risk and from moderaterisk to high-risk. The other three cases received a lower risk grading, where two cases were changed from high-risk to moderate-risk and one case was changed from high-risk to low-risk (Table 8).

Discussion
GIST is the most common mesenchymal tumor of the gastrointestinal tract. Most GISTs are activated due to mutations of c-kit proto-oncogene or platelet-derived growth factor receptor alpha gene. Its immunophenotype is positive for CD117, CD34, and DOG-1. The biological behavior of GIST has malignant potential and it is di cult to assess its prognosis accurately. The 2008 NIH modi ed assessment system is usually used to evaluate the postoperative recurrence risk of primary and resectable GISTs. The test parameters include primary location, tumor size, MI, and tumor rupture. Among these parameters, primary location, tumor size, and tumor rupture are easily established, while MI quanti cation is performed more subjectively and its repeatability is poor. However, MI determination in GIST or other soft tissue tumors is less repeatable among observers due to the diverse morphological manifestation of mitosis and maldistribution in tumors [4,5]. In addition, accurate quanti cation of mitosis on H&E-stained sections depends on the ability to locate "hot spot" regions with the highest mitosis rate and to identify true mitosis from apoptotic cells and nuclei fragments. Furthermore, this may underestimate MI in some cases and result in false assessment of biological behavior of GIST if the de nition of mitosis is too strict [6].
Therefore, it is necessary to identify a biomarker to represent the true status of the MI.
A histone is an important component of the nucleosome, which is a basic unit of chromatin in eukaryotes.
Researchers have identi ed ve types of histones, including H1, H2A, H2B, H3, and H4 [7]. Histones H2A, H2B, H3, and H4 can be phosphorylated under the effect of protein kinases. The modi ed sites of histone H3 protein are positions 3 and 11 of threonine and positions 10 and 28 of serine. In the process of mitosis, histone H3 begins to phosphorylate in the late G2 phase and diffuses to the entire chromosome as mitosis progresses. The phosphorylation reaches a peak level in the M phase, while dephosphorylation occurs at the end of the M phase. The phosphorylation degree of histone H3 is very low in the interphase, which means that phosphorylated histone H3 (PHH3) is mainly expressed in G2 and M phases [2,3]. At present, many studies consider that PHH3 is a speci c mitotic marker, which can help to distinguish mitosis, nuclear fragments, and apoptotic cells in tissue sections [8]. It can be easier and quicker to perform mitosis quanti cation objectively in meningiomas by detecting the expression status of PHH3. This test has important value for assessing meningioma prognosis [9][10][11][12]. Similarly, the histological grading of breast cancer has been changed by quantifying mitosis with PHH3 labeling. The PHH3 expression status is closely related to the prognosis of breast cancer [13][14][15][16]. In addition, MI quanti cation using MI-H&E has a strong correlation with PHH3-labeled MI in breast brous epithelial tumors, which can make MI quanti cation more convenient and is an important tool for diagnosis. In addition, researchers have discovered that PHH3-positive index has a signi cant correlation with tumor grading and prognosis in uterine smooth muscle tumor [17], astrocytoma [18][19], malignant melanoma [20][21], esophageal squamous cell carcinoma [22], ovarian serous adenocarcinoma [23], and other tumors.
Ki67 is a nuclear proliferation antigen that is expressed in all cell cycle phases, except for G0 phase, and is widely used to evaluate the proliferative activity of tumor cells. The Ki67-positive cells are not necessarily occurring in the M phase because Ki67 is expressed in all cell cycle phases, except for the G0 phase [24].
Although studies have shown that Ki67 is related to biological behavior of GIST, there is no system that uses it as an evaluation parameter for GIST grading [25][26][27][28]. Moreover, using it will result in overestimation of tumor proliferation activity because in ltrating lymphocytes in the tumor may also have an immunoreaction to Ki67.
The present study evaluated the relationship between MI-H&E and PHH3-positive index and clinicopathological characteristics of 125 cases of GIST. There were no statistically signi cant differences among MI-H&E, PHH3-positive index, and groups of different gender, age, primary location, and histological subtype. Correlation analysis showed that the PHH3-positive index has a good correlation with MI-H&E, which is consistent with previous research results [29][30]. Similarly, Ki67 index is correlated with MI-H&E.
In the 125 cases of GIST, both PHH3 and Ki67 had a high expression in 13 cases and low expression in 57 cases. Correlation analysis results showed that the Ki67 index was positively correlated with the PHH3positive index, but the correlation was lower than that between the PHH3-positive index and MI-H&E. This is due to the fact that Ki67 is expressed in all cell cycles except the G0 phase and Ki67 is also expressed by in ammatory cells in tumors. Comparatively, PHH3 is mainly expressed in the G2 and M phases and can represent cell mitosis status more accurately.
For a metastatic GIST in the present study, MI-H&E (31/5 mm 2 ) revealed a higher value than the PHH3positive index (1/5 mm 2 ). The tumor risk strati cation of this case has not been changed because the GIST was large enough in diameter (16 cm) to be graded as high-risk and therefore did not depend on the MI value for evaluation. However, correlation between MI-H&E and prognosis is better than that with the PHH3positive index in prognosis analysis of this special case. There are two possible reasons for this. First, false negative results during immunohistochemical staining may be due to various factors, such as insu cient tissue xation or improper operating procedure. Second, apoptotic cells or nuclear fragments affect the mitosis interpretation on H&E-stained sections. Our data show that the Ki67 index is related to GIST prognosis as well, which is consistent with previous research [25][26][27][28]. This suggests that both PHH3 and Ki67 are prognostic markers of GIST.
In the present study, tumor risk strati cation of ve cases was changed due to PHH3-IHC results. Compared to MI quanti cation using H&E-stained sections, two cases received a higher risk classi cation and three cases received a lower risk classi cation based on PHH3-IHC results. It is more di cult to accurately determine the "hot spot" regions and evaluate mitosis on H&E-stained sections due to interference from apoptotic cells and nuclear fragments [31]. This may lead to MI-H&E producing values that are higher than those for the PHH3-positive index, resulting in higher tumor risk strati cation using H&E-stained sections. By contrast, it is easy to identify the "hot spot" mitosis regions and distinguish real mitosis in a lowerpowered eld using PHH3 IHC staining. Therefore, it is more convenient and accurate to evaluate mitosis using PHH3-stained IHC sections. However, the false negative PHH3 IHC staining results may also affect mitosis quanti cation due to various factors [32][33][34][35]. Previous studies on PHH3 expression in malignant tumors have shown that PHH3-IHC is highly correlated with MI-H&E, although it is slightly higher than MI-H&E. However, only H&E-stained sections can be used to perform mitosis quanti cation during risk strati cation of GIST at this time [36][37][38]. For this reason, using PHH3-stained IHC sections to perform mitosis evaluation may lead to grading changes in tumor risk strati cation of GIST and affect further therapy. Accordingly, MI quanti cation using PHH3-stained IHC sections can be used as an assistive tool, rather than a replacement, for tumor risk strati cation of GIST when a pathological diagnosis is made.
In summary, PHH3-positive IHC index is positively correlated with MI-H&E, which is a more pronounced relationship than that between Ki67 in GIST. PHH3 can play an assistive role in evaluation of risk strati cation and prognosis in GIST. MI determined using PHH3-stained IHC sections can be used as an assistive tool, but not a replacement, for tumor risk strati cation of GIST using H&E-stained sections.

Conclusion
PHH3 IHC labeling is a potentially useful tool for risk strati cation and prognostic analysis in GIST. The use of PHH3 IHC labeling makes it more convenient for pathologists to determine the MI for GIST. MI quanti cation using PHH3 IHC sections can be used as an adjunct tool for risk strati cation and prognostic analysis of GIST, but cannot completely replace MI quanti cation using H&E-stained sections.
The Ki67 index is positively correlated with MI-H&E, but its e ciency for tumor risk strati cation is lower than that of PHH3.  Ki67 was positive in nuclei of GIST tumor cells (red arrow, ×40).