- 學位論文
二氫二醇去氫酶的過度表現作為可切除胃癌患者的預後指標
Overexpression of Dihydrodiol Dehydrogenase as a Prognostic Marker in Resected Gastric Cancer Patients
摘要
研究背景： 文獻上提到很多分子生物指標或使用組織病理上的一些特徵，像是淋巴血管浸潤侵犯情形來預測治療結果，都能得到不錯的準確度。有些則與胃癌有關，像是表皮生長因子接受體 ( Epidermal Growth Factor Receptor )，血管內皮生長因子( Vascular Endothelial Growth Factor )，缺氧誘發因子( Hypoxiainducible 1 alpha )，和P53基因等。從手術切除的胃癌標本，利用免疫組織化學染色法，分析二氫二醇去氫酶(DDH) 表現情形，由臨床病理資料，分析患者年齡、性別、腫瘤深度(T)、淋巴結轉移( N )、遠處轉移(M)、分期、組織學分類(WHO及 Lauren)及存活情形，結合兩者資料，是否可以DDH 表現情形預測胃癌之預後。 材料與方法： 從1998年1月到2004年9月，我們回溯性地收集81位在彰化基督教醫院外科接受胃切除的胃癌病人。對於DDH表現情形，病人的基本資料、組織學型態、分期及預後等之間相關性的研究，係採變異分析法(Analysis of variance)，定性資料彼此間相關性則採用chi-square test ，而各分群存活率的估算則採用Kaplan-Meier product-limit estimator，各分群存活率之間的比較則採用log-rank test。 研究結果： 81位胃癌病人，其基本資料詳見Table 5，五十五位男性及廿六位女性，平均年齡為65.1±12.3歲，平均追蹤6.2年(4.3~8.2年)，在這期間有四十六位病人死亡。腫瘤深度(T)以T2和T3居多，各分期佔率相差不多，組織學型態以管狀細胞居多。 利用免疫組織化學染色結果，有34位(42%)呈現DDH過度表現，47位(58%)為陰性。在存活組中陽性佔10位(28.5%)，而死亡組中陽性為24位(52%)，若以腫瘤侵犯深度(T)來分析，陽性率分別為T1:17%，T2:16%，T3:53%，T4:60%，按照淋巴結轉移資料分析，陽性率分別為N0:22%，N1:35%，N2:50%，N3:82% 。以病理分期(TNM) 來看，第I期陽性率為13% ，第II期爲24%，III期為52%，第IV為78%，從世界衛生組織(WHO)有關胃癌組織學型態分析，以戒指型細胞陽性率78% 最高，其次為分化不良型細胞48%，乳突狀細胞40%，管狀細胞36%最低。 由以上資料可知，DDH過度表現( 陽性率) 和TNM 分期(P=0.000)，腫瘤侵犯深度(P=0.018)，淋巴轉移(P=0.001) 在統計學上是有顯著差異的！而和性別，WHO 組織學型，Laurene 氏分類沒有相關性。 另外，使用Kaplan-Meier和 log-rank test 來分析比較五年活率以及各分組之間存活的差異性，而DDH低度表現（陰性反應）組有較好的存活率（P=0.048）。若再細分，可發現到胃癌期別（Stage）(P=0.001)，淋巴轉移（N因子）（P=0.002）有正相關性，與年齡、性別、WHO 組織學型態、Laurene 氏組織分類以及腫瘤侵犯深度無關。 討論： 從很多固態腫瘤（Solid tumor）可以偵測到DDH 的過度表現，像是肝細胞癌（Hepatocellular carcinoma ）、肺非小細胞癌（non-small cell lung cancer ）、卵巢癌和食道癌。更進一步可以發現到它會造成較差的預後，同時對於化學治療和放射治療有較高的抵抗力。然而它和胃腺癌的相性卻很少提及。本研究中，我們將所切除的胃癌組織標本，利用組織免疫染色法來分析DDH表現程度，合併分析臨床資料及預後，以期找出兩者的關係。 就我們所知，DDH在去毒化（Detoxification）過程，扮演一個很重要的角色。在肝臟aldo-keto reductase superfamily，DDH 是主要的酶，同時被用來代謝PAH。其中間代謝產物可和Glutathione或其他cellular substrates 相結合。癌細胞中，DDH 的過度表現和prostaglandins synthase activity增加，二者結合提供前列腺素的另一來源，進而影響調節細胞增生的功能。更進一步發現，癌細胞中 DDH 過度表現的情形比正常細胞來得多。 雖然在胃癌組織可偵測到DDH過度表現，但在正常胃部組織中並不會常態性地看到這個現象。所以其中的誘發（Induction） 機轉仍然不明！在本研究中，胃癌組織中有DDH表現者，可見到較差的預後，若呈現過度表現者，則和高死亡率，TNM第四期及腫瘤深度為T3或T4有相關性。 81位病人中，DDH 表現率為４１％，其與TNM 分期有關，而與Lanren 氏組織型態中的分化不良者無關。値得注意的是DDH 過度表現者多見於Advanced gastric cancer，尤其在腫瘤深度侵犯漿膜層(T3)，腫瘤深度侵犯鄰近器官，以及淋巴結轉移數超過十五個（N3）更是有明顯差異！它同時也是術後影響存活率的一個指標，但是它如何影響胃癌的臨床表現仍須有待進一步的研究。另外，本篇研究較少有早期的胃癌組織標本，例如腫瘤只侵犯粘膜或肌肉層者（T1，T2），和較少淋巴結轉移者（N0，N1）, 無法做很好的對照組。原因在於本院早期胃癌比例為15%，遠低於日、韓（約50%），個案收集上較困難。 就我們所知，這是第一篇以胃癌切除標本來研究DDH表現程度與存活率的文章，我們的結果顯示，若有DDH陽性反應者，有較差的預後，術後積極施以輔助治療（如化學治療或放射線治療），也許能提高存活率。
並列摘要
Background and subjective： Gastric cancer remains a major health hazard internationally especially in the Far East, including southern China and Taiwan. Survival over the last 20 years has improved, not only in Asia, but also in the West. Use of histopathological features, for example lymphovascular invasion or molecular markers, for prognosis may improve survival. Several molecular markers have been associated with prognosis of gastric cancer, including epidermal growth factor receptor, vascular endothelial growth factor, hypoxia-inducible factor 1alpha, and p53. Recently, overexpression of dihydrodiol dehydrogenase (DDH) was shown in specimens of non-small-cell lung cancer, ovarian carcinoma, and hepatocellular carcinoma. Incidence of recurrence and distant metastasis was found to be significantly higher in patients with high tumor DDH than in those with low DDH expression in human ovarian carcinoma cells. Moreover, survival was significantly better in patients with low DDH expression. The cytoplasmic enzyme DDH plays an important role in detoxification. DDH is a member of the aldo-keto reductase superfamily, which normally reduces polycyclic aromatic hydrocarbons (PAH) in the liver. Four isoforms of DDH (DDH1–DDH4) have been identified (each with monomeric mass 36 kDa). DDH overexpression indicates an alternative pathway of prostaglandin synthesis. The resulting increase in prostaglandin may explain the chronic inflammation associated with carcinogenesis of the stomach, and, possibly, other manifestations of gastric cancer. In this study, we evaluated the prognostic role of DDH expression in a consecutive series of 81 gastric cancer patients treated with curative surgical resection. We also analyzed the correlation between DDH expression, clinicopathological features, and survival. Material and methods： The Ethics Committee of Changhua Christian Hospital approved the protocol of this study, and written informed consent was obtained from every patient or family. Between January 1998 and September 2004, we retrospectively enrolled 81 patients with gastric cancer admitted to the Department of General Surgery of Changhua Christian Medical Center in central Taiwan. The inclusion criteria were: 1 tumor location in the stomach, and 2 complete resection of gastric cancer. All patients had pathologically proven adenocarcinoma of the stomach by endoscopic biopsy. All patients received preoperative endoscopic ultrasonic examination and abdominal computerized tomography for clinical staging. Three specimens from each tumor were paraffin-embedded. The tissue preparation and staining procedure have been described previously. Briefly, after 4 mm sections were cut, the wax was melted in a 65_C oven overnight. The slides were deparaffinized in xylene, which was removed subsequently with absolute ethanol before immunostaining. The DDH polyclonal antibody was reactive to the four isoforms of DDH (DDH1, DDH2, DDH3, and DDH4; Cashmere Scientific Company, Taipei, Taiwan). The slides were incubated with monoclonal antibody to DDH and then with peroxidase-conjugated goat antimouse immunoglobulin. Diaminobenzidine was used as chromogenic substrate, and brown precipitate was identified as positive staining. The sections were counterstained with hematoxylin and mounted in glycerol gelatin. Each group of slides included a positive and a negative control. Under low-power, four areas containing tumor cells on each slide were selected at random and photographed for evaluation. Each slide was read and scored by two independent investigators. The staining intensity was scored semiquantitatively according to percentage of positively stained cells. Low expression was defined as less than 10% positive cells. Overexpression was defined as moderate staining (10–50% positive cells) or strong staining (more than 50% positive cells). According to the intensity of staining per cell, low expression was defined as very low cell intensity staining. Overexpression was defined as intermediate or very high intensity staining. Relationships between metastasis, expression of DDH, and other quantitative data were studied by analysis of variance. Relationships among qualitative data were analyzed using the chi-square test (or two-tailed Fisher exact test when the expected number of any cell was smaller than five cases). The Kaplan–Meier product-limit estimator was used to estimate cancer-specific survival for subgroups with or without DDH expression. The log-rank test was used to compare these subgroups with respect to cancerspecific survival. Result： This study enrolled 81 patients (55 men and 26 women; mean age, 65.1 ± 12.3 years; 29–82 years). The median follow-up time was 6.2 years (range 4.3–8.2 years). During the follow-up period, 46 patients died and 35 survived. DDH overexpression occurred in 41.9% of gastric cancer patients. Overexpression rate was 28.5% among survivors and 52.1% among nonsurvivors. The difference in DDH expression level between the high and low expression groups of nonsurvivors was statistically significant (P = 0.042) by the chi-square test. TMN stage, tumor classification, and node classification also differed significantly between the high and low DDH expression groups. The tumors for 34 patients (41.9%) were positive for DDH overexpression. Of these patients, 13% had stage I, 24% had stage II, 52% had stage III, and 78% had stage IV tumors. Significant difference in clinicopathological data, for example TMN stage (P = 0.000), tumor classification (P = 0.018), and node classification (P = 0.001) but not in gender, WHO classification, and Lauren’s classification was found between groups with and without DDH overexpression . Interestingly, the five-year survival rate was also significantly better for the low DDH expression group than the high DDH expression group (P = 0.048) using Kaplan– Meier analysis with the log-rank tes0074. For tumor stage (P = 0.001) and node classification (P = 0.002) but not age, gender, WHO classification, and Lauren’s classification, five-year survival was significantly correlated with DDH expression. Although five-year survival was more likely in patients with low DDH expression, the statistical difference was weakly significant (P = 0.048). Discussion： DDH overexpression has been detected in the majority of solid tumors, for example hepatocellular carcinoma, nonsmall-cell lung cancer, ovarian cancer, and esophageal cancer . Furthermore, in several cancers, DDH overexpression has been associated with poor prognosis and resistance to chemotherapy or radiation therapy . However, the clinical significance of DDH status in gastric adenocarcinoma has not been reported. In this study, we examined DDH expression in surgical specimens of gastric cancer using immunohistochemistry, and its clinical significance was evaluated. As noted above, DDH contributes to detoxification. DDH is a major enzyme of the aldo-keto reductase superfamily in the liver and used to metabolize PAH. The intermediate metabolites conjugate to glutathione or the other cellular substrates. The association of DDH expression with increasing prostaglandins synthase activity in cancer cells suggests that another source of prostaglandins may affect the regulation of cell proliferation. Furthermore, the increase in DDH expression may be more marked in cancer cells. Although detected in gastric cancer, DDH is not regularly expressed in the normal human stomach. Thus, the induction mechanism of DDH overexpression remains to be clarified. In our series, DDH expression in resected gastric cancer specimens correlated with mortality, TMN stage, and tumor classification. DDH overexpression was significantly more frequent in patients who died of gastric cancer, or patients with TMN stage IV, and tumor classifications T4 and N3. In this study, 41.9% of 81 specimens were DDH-positive, correlating with TNM staging but not poorly differentiated histology (Lauren’s type). This finding suggested that cancer progression is associated with increased DDH expression. Notably, the DDH overexpression rate was 54% in patients with advanced (T3 and T4) disease and 82% in patients with N3 disease. Like advanced tumor extension (T4) and the presence of nodal metastases (N3), DDH expression in our study was an independent prognostic indicator of clinical outcome in gastric cancer patients undergoing potentially curative surgery. We considered that knowledge of the molecular pathways determining the behavior of individual gastric tumors may be needed for identifying high-risk patients. Furthermore, our study did not include enough patients with early-stage gastric cancers (T1–T2, N0-2), which are associated with good long-term prognosis, to prove the correlation between the DDH overexpression and clinical outcome. Estimation of the five-year survival probability showed that patients with low gastric cancer DDH expression had good outcome. To our knowledge, this is the first study demonstrating that, in gastric cancer patients undergoing potentially curative surgery, DDH overexpression correlates with poor long-term prognosis. Thus, the high-risk subset of patients with cancers in the same stage can be identified. Our results suggest that DDH positivity may be used as a poor prognosis marker to select candidates for adjuvant radiotherapy or chemotherapy following potentially curative surgery for early-stage disease. A substantial proportion of gastric cancer patients undergoing potentially curative surgery ultimately develop recurrent disease. In the past few years, new molecular markers of apparently homogeneous disease have been sought that are able to select patients with different prognosis and with different probabilities of benefiting from treatment. This study suggests that DDH overexpression may be a useful marker in identifying eligibility of highrisk gastric cancer patients for multimodal treatments. However, a more extensive and detailed study is required to prove the correlation between DDH overexpression and clinical outcome in patients with early-stage gastric cancer.