Prognostic and therapeutic implications of lysyl oxidase and cyclooxygenase 2 expressions in epithelial ovarian carcinoma

Patients

This study was performed in the Departments of Obstetrics and Gynecology, Pathology and Clinical Oncology, Faculty of Medicine, Zagazig University and included 70 patients with epithelial ovarian tumors who were registered in the period from January 2012 to December 2016. Clinicopathological characteristics of the patients were obtained. Patients were classified according to criteria of the International Federation of Gynecology and Obstetrics (FIGO) staging system [12]. This study was conducted in accordance with the Declaration of Helsinki and written consent was obtained from each participant. Also, the present study was permitted by the Ethical Committee of our university.

 

Regarding tumor histology and differentiation degree, 70 blocks of formalin-fixed-paraffin-embedded ovarian tissue (one for each case) were evaluated. Cases diagnosed with EOC were examined by two independent pathologists, classified and graded according to the World Health Organization (WHO) classification system [13]. Other types of ovarian cancers rather than EOC were excluded. All patients were followed up for survival data and the mean follow up period was three years. Regarding patients with EOC, clinical follow-up was done every three months to all cases, and patients who received any form of chemotherapy were assessed for response according to Response Evaluation Criteria in Solid Tumors (RECIST) [14]. Among the 40 EOC patients, only 14 patients with the resectable early- stage disease were treated with radical surgery and adjuvant chemotherapy. In addition, three patients with the initial advanced-stage disease who achieved a complete response to neoadjuvant chemotherapy followed by interval cytoreductive surgery. These patients were followed and assessed for the presence or absence of relapse.

 

Immunohistochemistry (IHC)

The immunohistochemical staining procedure was done using the streptavidin-biotin immunoperoxidase technique (Dako-Cytomation, Glostrup, Denmark) [15]. Sections of 3–5 μm from the formalin-fixed-paraffin-embedded blocks were cut and mounted on positively charged slides then deparaffinized by xylene and rehydrated in graded alcohol. Thereafter, sections were boiled in buffered citrate (pH 6.0) for about 20 minutes then washed in phosphate-buffered saline (PBS) (pH 7.3). Then, endogenous peroxidase activity was blocked with 6% hydrogen peroxide in methanol. Slides from all blocks were incubated overnight with rabbit polyclonal COX2 (Thermo scientific, Fermont, USA, 1:50), others were incubated with mouse monoclonal LOX (GeneTex International Corporation GTX84182, Dilution 1:50) antibodies. After rinsing in PBS, the slides were immersed with a biotin-conjugated secondary antibody (Lab Vision Corporation, Fermont, USA). DAB was used as a chromogen and Mayer's Hematoxylin was used as a counterstain and then the slides were washed with distilled water and PBS. Positive and negative controls were stained with the same setting of the studied cases. Lung carcinoma and breast cancer tissue sections were used as a positive control for COX2 and LOX respectively. For negative control, primary antibodies were replaced by non-immune normal serum.

 

Immunohistochemical scoring

Cytoplasmic stain as positive for COX has been considered positive and the distribution of COX2 was scored independently and in blinded manner by two investigators on the following scale; 0, no staining; 1, weak cytoplasmic staining (may contain stronger intensity in <10% of the cancer cells); 2, moderate cytoplasmic staining in >10% of the cancer cells; and 3, strong cytoplasmic staining in >50% of cancer cells [16].

 

Nuclear staining has been considered positive for LOX and the scoring criteria for LOX intensity were: no color, 0; light yellow, 1; yellow, 2; and brown, 3. For the percentage of tumor positivity, the following scoring was used: negative, 0; 1-25%, 1; 25-50%, 2; and >50%, 3. Both the staining intensity and percentage positivity scores were summed and tumors with scores ranging from 0 to 9 were assigned to: negative (0-1), + (2), ++ (3-4), +++ ≥5 [17].

 

Statistical analysis

Continuous variables were expressed as the (mean±SD) and median (range) and the categorical variables were expressed as a number (percentage). The continuous variables were checked for normality by using the Shapiro-Wilk test. Mann Whitney U test was used to compare between two groups of non-normally distributed variables. Kruskal Wallis H test was used to compare between more than two groups of normally distributed variables. Percent of categorical variables were compared using Pearson's Chi-square test or Fisher's exact test when appropriate. Disease Free Survival (DFS) was calculated as the time from date of surgery to relapse or the most recent follow- up in which no relapse was detected. Overall survival (OS) was calculated as the time from diagnosis to death or the most recent follow- up contact (censored). Stratification of DFS and OS was done according to immunohistochemical markers. These time-to-event distributions were estimated using the method of Kaplan-Meier plot and compared using two-sided exact log-rank test. A p-value of ≤0.05 was considered significant. All statistics were performed using SPSS 22.0 for Windows (SPSS Inc., Chicago, IL, USA) and MedCalc Windows (MedCalc Software bvba 13, Ostend, Belgium).

 Table 1. Clinicopathological features, immunohistochemical markers of 70 patients with ovarian tumors

	All (N=70)
	N	(%)
Age (years)
Mean ± SD	46.11	±11.02
Median (Range)	45	22-67
≤50 years	44	62.9%
>50 years	26	37.1%
Tumor type
Benign tumor	20	28.6%
Borderline tumor	10	14.3%
Malignant tumor	40	57.1%
Histological type
Serous type	51	72.9%
Mucinous type	15	21.4%
Endometrioid type	4	5.7%
COX2
Negative	27	38.6%
Positive	43	61.4%
LOX
Negative	33	47.1%
Positive	37	52.9%
COX2/LOX
Negative/Negative	22	31.4%
Negative/Positive	5	7.1%
Positive/Negative	11	15.7%
Positive/Positive	32	45.7%

 

Table 2. The comparison between benign, borderline and malignant ovarian tumors

	All     (N=70)		Benign ovarian tumors (N=20)		Borderline ovarian tumors (N=10)		Malignant ovarian tumors (N=40)		p-value
	N	(%)	N	(%)	N	(%)	N	(%)
Age (years)
Mean ± SD	46.11	±11.02	38.50	±8.40	41.80	±11.39	51	±9.57	<0.001•
Median (Range)	45	22-67	39.50	22-55	39.50	22-63	52.50	28-67
≤50 years	44	62.9%	19	95%	8	80%	17	42.5%	<0.001‡
>50 years	26	37.1%	1	5%	2	20%	23	57.5%
Histological type
Serous type	51	72.9%	15	75%	7	70%	29	72.5%	0.437‡
Mucinous type	15	21.4%	5	25%	3	30%	7	17.5%
Endometrioid type	4	5.7%	0	0%	0	0%	4	10%
COX2
Negative	27	38.6%	14	70%	5	50%	8	20%	0.001‡
Positive	43	61.4%	6	30%	5	50%	32	80%
LOX
Negative	33	47.1%	15	75%	6	60%	12	30%	0.003‡
Positive	37	52.9%	5	25%	4	40%	28	70%
COX2/LOX
Negative/Negative	22	31.4%	12	60%	3	30%	7	17.5%	0.001‡
Negative/Positive	5	7.1%	2	10%	2	20%	1	2.5%
Positive/Negative	11	15.7%	3	15%	3	30%	5	12.5%
Positive/Positive	32	45.7%	3	15%	2	20%	27	67.5%

•  Kruskal Wallis H test, ‡ Chi-square test.

Patients’ characteristics

The present study included 70 patients with epithelial ovarian tumors: 20 benign, 10 borderline and 40 EOC. The age distribution of patients with benign tumors ranged between 22-55 years with a mean age of 38.5 (±8.4) while the age of patients with borderline tumors ranged between 22-63 years with a mean age of 41.8 (±11.39) (Tables 1, 2).

 

Regarding EOC patients, the age at the time of initial diagnosis ranged from 28 to 67 years with a mean age of 51 (±9.57) years. Fourteen of these 40 patients (35%) presented with early-stage disease and underwent radical surgery followed by adjuvant chemotherapy. On the other hand, twenty-six of EOC patients (65%) presented with advanced bulky disease or with significant comorbidities which precluded primary debulking surgery. These patients had received adjuvant chemotherapy and evaluated for response according to RECIST criteria. The median follow-up time was 26.5 months (range: 12-36 months) and during the follow-up period, 29.4% of the patients were disease- free without relapse. Recurrence occurred in 70.6% of the patients, while 62.5% of the patients died during follow-up. The clinicopathological characteristics of the 40 patients with EOC are summarized in Table 3

 

Table 3. Clinicopathological features, immunohistochemical markers and outcome of 40 patients with malignant ovarian tumors

Characteristics	All (N=40)		Characteristics	All (N=40)
	N	(%)		N	(%)
Age (years)			LOX
Mean ± SD	51	±9.57	Negative	12	30%
Median (Range)	52.50	28-67	Positive	28	70%
≤50 years	17	42.5%	COX2/LOX
>50 years	23	57.5%	Negative/Negative	7	17.5%
Histological type			Negative/Positive	1	2.5%
Serous type	29	72.5%	Positive/Negative	5	12.5%
Mucinous type	7	17.5%	Positive/Positive	27	67.5%
Endometrioid type	4	10%	Response	(N=26)
Grade			Complete response	3	11.5%
Grade I	7	17.5%	Partial response	2	7.7%
Grade II	18	45%	Stationary disease	7	26.9%
Grade III	15	37.5%	Progressive disease	14	53.8%
Distant metastasis			Overall response	5	19.2%
Absent	27	67.5%	No response	21	80.8%
Present	13	32.5%	Follow-up duration (months)
Stage			Mean ± SD	26.72	±8.60
Stage I	5	12.5%	Median (Range)	26.50	12-36
Stage II	9	22.5%	Relapse	(N=17)
Stage III	13	32.5%	Absent	5	29.4%
Stage IV	13	32.5%	Present	12	70.6%
COX2			Mortality
Negative	8	20%	Alive	15	37.5%
Positive	32	80%	Died	25	62.5%

Table 4. The relation between clinicopathological features and immunohistochemical markers of 40 patients with malignant ovarian tumors

Characteristics	All (N=40)		COX2					p-value	LOX					p-value
			Negative (N=8)			Positive (N=32)			Negative (N=12)			Positive (N=28)
	N	(%)	N	(%)		N	(%)		N	(%)		N	(%)
Age (years)
Mean ± SD	51	±9.57	44.50	±8.94		52.62	±9.14	0.050•	48.58	±9.84		52.03	±9.44	0.360•
Median (Range)	52.50	28-67	46	28-56		54	35-67		51	28-65		53.50	35-67
≤50 years	17	42.5%	6	35.3%		11	64.7%	0.053‡	6	35.3%		11	64.7%	0.530‡
>50 years	23	57.5%	2	8.7%		21	91.3%		6	26.1%		17	73.9%
Histological type
Serous type	29	72.5%	4	13.8%		25	86.2%	0.195‡	7	24.1%		22	75.9%	0.409‡
Mucinous type	7	17.5%	2	28.6%		5	71.4%		3	42.9%		4	57.1%
Endometrioid type	4	10%	2	50%		2	50%		2	50%		2	50%
Grade
Grade I	7	17.5%	6	85.7%		1	14.3%	<0.001§	6	85.7%		1	14.3%	<0.001§
Grade II	18	45%	2	11.1%		16	88.9%		6	33.3%		12	66.7%
Grade III	15	37.5%	0	0%		15	100%		0	0%		15	100%
Distant metastasis
Absent	27	67.5%	8	29.6%		19	70.4%	0.037‡	12	44.4%		15	55.6%	0.004‡
Present	13	32.5%	0	0%		13	100%		0	0%		13	100%
Stage
Stage I	5	12.5%	5	100%		0	0%	<0.001§	4	80%		1	20%	<0.001§
Stage II	9	22.5%	3	33.3%		6	66.7%		5	55.6%		4	44.4%
Stage III	13	32.5%	0	0%		13	100%		3	23.1%		10	76.9%
Stage IV	13	32.5%	0	0%		13	100%		0	0%		13	100%
COX2
Negative	8	20%							7	87.5%		1	12.5%	<0.001‡
Positive	32	80%							5	15.6%		27	84.4%
LOX
Negative	12	30%	7	58.3%		5	41.7%	<0.001‡
Positive	28	70%	1	3.6%		27	96.4%

 •  Mann Whitney U test, ‡ Chi-square test, § Chi-square test for trend.

Association of LOX and COX2 expression with clinicopathological parameters

Regarding EOC patients, positive LOX expression was observed in 70% of the patients and positive COX2 IHC staining was observed in 80% of the patients (Tables 4, 5). COX2 was stained in the cytoplasm of cancer cells (Figure 1) while LOX was stained in the nuclei of cancer cells (Figure 2). Both biomarkers were significantly correlated with tumor grade (p<0.001 for both). In addition, a significant association was observed between FIGO stage and LOX IHC staining, with 100% of the patients with stage IV disease having positive staining versus 20% of the patients with stage I disease (p<0.001). Furthermore, a significant association between FIGO stage and COX2 was observed, in which 100% of the patients with stage IV disease had positive staining versus 0% of the patients with stage I disease. In the same way, a significant association between FIGO stage and LOX/COX2 IHC staining was observed, in which 100% of the patients with stage IV had positive staining versus 0% of the patients with stage I disease. Moreover, there was a significant association between LOX and COX2 IHC staining, in which 96.4% of the patients with positive staining for LOX showed positive staining for COX2 versus 58.3% of the patients with negative staining for LOX showed negative staining for COX2.

Characteristics	All (N=40)		COX2/LOX											p-value
			Negative/Negative (N=7)			Negative/Positive (N=1)			Positive/Negative (N=5)			Positive/Positive (N=27)
	N	(%)	N	(%)		N	(%)		N	(%)		N	(%)
Age (years)
Mean ± SD	51	±9.57	44.14	±9.59		47			54.80	±6.76		52.22	±9.56	0.237•
Median (Range)	52.50	28-67	45	28-56					54	46-65		54	35-67
≤50 years	17	42.5%	5	29.4%		1	5.9%		1	5.9%		10	58.8%	0.164‡
>50 years	23	57.5%	2	8.7%		0	0%		4	17.4%		17	73.9%
Histological type
Serous type	29	72.5%	3	10.3%		1	3.4%		4	13.8%		21	72.4%	0.533‡
Mucinous type	7	17.5%	2	28.6%		0	0%		1	14.3%		4	57.1%
Endometrioid type	4	10%	2	50%		0	0%		0	0%		2	50%
Grade
Grade I	7	17.5%	5	71.4%		1	14.3%		1	14.3%		0	0%	<0.001§
Grade II	18	45%	2	11.1%		0	0%		4	22.2%		12	66.7%
Grade III	15	37.5%	0	0%		0	0%		0	0%		15	100%
Distant metastasis
Absent	27	67.5%	7	25.9%		1	3.7%		5	18.5%		14	51.9%	0.026‡
Present	13	32.5%	0	0%		0	0%		0	0%		13	100%
Stage
Stage I	5	12.5%	4	80%		1	20%		0	0%		0	0%	<0.001§
Stage II	9	22.5%	3	33.3%		0	0%		2	22.2%		4	44.4%
Stage III	13	32.5%	0	0%		0	0%		3	23.1%		10	76.9%
Stage IV	13	32.5%	0	0%		0	0%		0	0%		13	100%

 •  Kruskal Wallis H test, ‡ Chi-square test, § Chi-square test for trend.

Table 6. The relation between each immunohistochemical marker and outcome of 40 patients with malignant ovarian tumors

Characteristics	All		COX2					p-value	LOX					p-value
			Negative			Positive			Negative			Positive
	N	(%)	N	(%)		N	(%)		N	(%)		N	(%)
Response	(N=26)					(N=26)			(N=3)			(N=23)
Complete response	3	11.5%				3	11.5%		3	100%		0	0%	<0.001‡
Partial response	2	7.7%				2	7.7%		0	0%		2	8.7%
Stationary disease	7	26.9%				7	26.9%		0	0%		7	30.4%
Progressive disease	14	53.8%				14	53.8%		0	0%		14	60.9%
Overall response	5	19.2%				5	19.2%		3	100%		2	8.7%	0.004‡
No response	21	80.8%				21	80.8%		0	0%		21	91.3%
Relapse	(N=17)		(N=8)			(N=9)			(N=12)			(N=5)
Absent	5	29.4%	4	50%		1	11.1%	0.131‡	5	41.7%		0	0%	0.245‡
Present	12	70.6%	4	50%		8	88.9%		7	58.3%		5	100%
Disease free survival
Mean (month) (95% CI)	30.12 months (27.58-32.66)		34 months (32.45-35.55)			26.67 months (23.45-29.88)		0.006†	32.17 months (29.87-34.46)			25.20 months (20.46-29.94)		0.013†
Median (months)	31 months		34 months			26 months			32 months			24 months
2-year DFS	76.5%		100%			55.6%			91.7%			40%
3-year DFS	29.4%		50%			11.1%			41.7%			0%
Mortality	(N=40)		(N=8)			(N=32)			(N=12)			(N=28)
Alive	15	37.5%	8	100%		7	21.9%	<0.001‡	12	100%		3	10.7%	<0.001‡
Died	25	62.5%	0	0%		25	78.1%		0	0%		25	89.3%
Overall survival
Mean (month) (95% CI)	26.73 months (24.09-29.36)		36 months			24.41 months (21.65-27.17)		<0.001†	36 months			22.75 months (20.12-25.38)		<0.001†
Median (months)	24 months		Not reached			24 months			Not reached			22 months
1-year OS	95%		100%			93.8%			100%			92.9%
2-year OS	50%		100%			37.5%			100%			28.6%
3-year OS	37.5%		100%			21.9%			100%			10.7%

 ‡ Chi-square test, † Log rank test.

 

Association between LOX and COX2 expression and response to adjuvant chemotherapy

Among the twenty-six patients who were inoperable at the time of diagnosis (either due to advanced bulky disease or presence of comorbidity), three patients (11.5%) had complete recovery (CR), two patients (7.7%) had partial recovery (PR), seven patients (26.9%) had stationary disease (SD) and fourteen (53.8%) patients had progressive disease (PD). LOX/COX2 (positive/positive) staining was significantly associated with poor response to chemotherapy (p<0.001) (Tables 6, 7).

 

Association between LOX and COX2 expression and survival

We analyzed the disease free survival (DFS) and overall survival (OS) of EOC patients using the Kaplan-Meier method (Figure 3). Regarding DFS, a significant difference existed between patients with negative versus those with positive LOX expressions. The mean DFS for patients with negative expression was significantly longer than that with positive expression (32.17 versus 25.2 months, p≤0.013) and the 3-year DFS was 41.7 % versus 0%, respectively. With respect to mortality rate, a significant association between LOX IHC staining was observed, where 89.3% of the patients with positive staining died versus 0% of the patients with negative staining (p≤0.001). Similarly, a significant association between COX2 IHC staining and mortality was observed, in which 78.1% of the patients with positive staining died versus 0% of patients with negative staining.

 

Moreover, a significant difference was observed between patients with negative LOX expression and those with positive expression with respect to OS, where the mean OS for negative patients was significantly longer than that of positive patients (36 versus 22.75 months, p<0.001) and 3-year OS was 100% versus 10.7%, respectively. In addition, a significant difference between negative COX2 and positive COX2 patients with respect to OS, where the mean OS for negative patients was significantly longer than the mean OS for positive patients (36 versus 24.41 months, p<0.001) and the 3-year OS was 100% versus 21.9%, respectively. Furthermore, a significant association between LOX/COX2 IHC staining and mortality was found, in which 92.6% of the patients with positive/positive staining died versus 0% of patients with negative/negative staining (Figure 3, Tables 6, 7).

It has been established that ovarian cancer development reflects a loss of tissue organization and differentiation. These processes have been associated with increases in LOX expression and activity that regulate EMT transcription factors [11].

 

                                                          

 

                                                            

 

Figure 3. Kaplan Meier Survival Plots, Left Panel: Disease Free Survival (DFS); Right Panel: Overall Survival, (A) and (E): All studied malignant ovarian tumor patients, (B) and (F) Stratified by COX2 IHC staining, (C) and (G) Stratified by LOX IHC staining, (D) and (H) Stratified by COX2/LOX IHC staining.

 

Given the potential link of LOX to specific localization, De Donato et al [17] assessed only the nuclear expression of LOX in ovarian tumors, while Ji et al investigated mainly its cytoplasmic expression [18]. In this study, we examined the nuclear expression of LOX in EOC. LOX overexpression was detected in EOC tumor tissues compared with non-cancer tissues with a statistically significant difference which agree with other studies [18]. Many cancers showed LOX overexpression compared with their normal or non-aggressive neoplasms [19,20].

 

It has been documented that LOX protein structure and function are so complex and involve vital biological processes, such as cell movement, signal transduction and gene regulation [21]. Our results revealed a positive relationship between LOX expression and grade of malignancy. This is consistent with other studies [18,22]. Polymorphism of LOX was associated with increased risk of EOC and promoted the ovarian cancer progression and correlated with a lower tumor differentiation degree and with lymph node metastasis [22]. On the other hand, Wang et al reported that overexpression of LOX suppressed the proliferation and invasion of ovarian cancer cells, demonstrating the tumor suppressor function of the LOX gene [21].

 

LOX plays a critical role during the formation of premetastatic niches by stimulating collagen cross-linking and fibronectin synthesis, facilitating tumor-stromal interactions that are important for tumor progression and metastasis [10]. This role has been extensively studied in breast cancer in which LOX was considered to be important for late-stage tumor progression to metastasis [19]. Our results were consistent with Ji et al [18] who reported that LOX expression was positively correlated with advanced stage. These results demonstrated that LOX overexpression is an independent prognostic factor of a worse outcome in EOC patients.

Regarding the response to chemotherapy, our findings are similar to that of De Donato et al who reported an association between LOX expression and chemotherapy resistance [17]. LOX activity modulates the physical barrier function of the ECM for small molecule drugs thus influencing their therapeutic efficacy [23]. Therefore, therapeutic targeting of LOX may one day improve the clinical course of metastatic ovarian cancer patients.

 

With respect to survival analysis, we found that EOC patients with high LOX expression had a poor outcome as indicated by low rates of OS and DFS. This is in accordance with other studies which demonstrated that LOX overexpression was significantly associated with decreased OS (p=0.009) and DFS (p=0.048) in EOC [24]. However, there is still some debate regarding the association of LOX overexpression and the pathogenesis of invasive carcinomas [25]. On one hand, we have increased LOX expression in several types of cancer which is considered an independent prognostic marker of a worse outcome in gastric cancer and in lung adenocarcinoma [26]. On the other hand, LOX appears to act as a tumor suppressor in other cancers such as human osteosarcoma [27]. Therefore, the suggestion of whether LOX enhances or suppresses tumor progression in various tumors is still controversial. An explanation for such seemingly paradoxical role for LOX in cancer is likely the existence of multiple forms and the differential localization of LOX. Moreover, LOX expression may vary during the different stages of transformation and with different tissue types [28].

 

The COX2 has been established to be frequently expressed in human adenocarcinomas and its inhibition suppresses tumor progression [16]. In the present study, COX2 expression was detected in 80% of EOC cases which differed significantly from benign adenoma and borderline tumors (p=0.001). These results are consistent with other studies [7,16] where COX2 positive EOC cases were significantly correlated with poor differentiation, advanced stage, short survival time and resistance to chemotherapy as compared to patients with tumor stained negative for COX2.

1. Nassar HR, Zeeneldin AA, Helal AM, Ismail YM, Elsayed AM, Elbassuiony MA, et al. Treatment outcomes of epithelial ovarian cancers following maximum cytoreduction and adjuvant paclitaxel-carboplatin chemotherapy: Egyptian NCI experience. Asian Pac J Cancer Prev 2015;16:7237-42.
2. Thomassin-Naggara I, Bazot M, Daraï E, Callard P, Thomassin J, Cuenod CA. Epithelial ovarian tumors: value of dynamic contrast-enhanced MR imaging and correlation with tumor angiogenesis. Radiology 2008;248:148-59.
3. Siegel R, Naishadham D, Jemal A. Cancer statistics. CA Cancer J Clin 2013;63:11-30.
4. Heintz AP, Odicino F, Maisonneuve P, Quinn MA, Benedet JL, Creasman WT, et al. Carcinoma of the ovary. FIGO 26th annual report on the results of treatment in gynecological cancer. Int J Gynaecol Obstet 2006;95 Suppl 1:S161-92.
5. Ozols RF, Bundy BN, Greer BE, Fowler JM, Clarke-Pearson D, Burger RA, et al. Phase III trial of carboplatin and paclitaxel compared with cisplatin and paclitaxel in patients with optimally resected stage III ovarian cancer: A Gynecologic Oncology Group study. J Clin Oncol 2003;21:3194-200.
6. Meyer S, Vogt T, Landthaler M, Berand A, Reichle A, Bataille F, et al. Cyclooxygenase 2 (COX2) and Peroxisome Proliferator-Activated Receptor Gamma (PPARG) are stage-dependent prognostic markers of malignant melanoma. PPAR Res 2009;2009:848645.
7. Rask K, Zhu Y, Wang W, Hedin L, Sundfeldt K. Ovarian epithelial cancer: a role of PGE2-synthesais and signaling in malignant transformation and progression. Mol Cancer 2006;5:62.
8. Mutoh M, Takahashi M, Wakabayashi K. Roles of prostanoids in colon carcinogenesis and their potential targeting for cancer chemoprevention. Curr Pharm Des 2006;12:2375-82.
9. Mayorca-Guiliani A, Erler JT. The potential for targeting extracellular LOX proteins in human malignancy. Onco Targets Ther 2013;6:1729–35.
10. Taylor MA, Amin JD, Kirschmann DA, Schiemann WP. Lysyl oxidase contributes to Mechanotransduction-Mediated Regulation of Transforming Growth Factor-β signaling in breast cancer cells. Neoplasia 2011;13:406-18.
11. Kim K, Park EY, Yoon MS, Suh DS, Kim KH, Lee JH, et al. The role of TWIST in ovarian epithelial cancers. Korean J Pathol 2014;48: 283-291.
12. Prat J. FIGO Committee on Gynecologic Oncology. Staging classification for cancer of the ovary, fallopian tube, and peritoneum. Int J Gynaecol Obstet 2014;124:1-5.
13. Kurman RJ, Carcangiu ML, Herrington CS, Young RH. WHO classification of tumors of female reproductive organs. 4th ed. Lyon: International Agency for Research on Cancer; 2014.
14. Eisenhauer EA, Therasse P, Bogaerts J, Schwartz LH, Sargent D, Ford R, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 2009;45:228-47.
15. Hsu SM, Raine L, Fanger H. Use of avidin-biotinperoxidase complex (ABC) in immunoperoxidase techniques: a comparison between ABC and unlabeled antibody (PAP) procedures. J Histochem Cytochem 1981;29:577-80.
16. Erkinheimo TL, Lassus H, Finne P, van Rees BP, Leminen A, Ylikorkala O, et al. Elevated cyclooxygenase-2 expression is associated with altered expression of p53 and SMAD4, amplification of HER-2/neu, and poor outcome in serous ovarian carcinoma. Clin Cancer Res 2004;10:538–45.
17. De Donato M, Petrillo M, Martinelli E, Filippetti F, Zannoni GF, Scambia G, et al. Uncovering the role of nuclear Lysyl oxidase (LOX) in advanced high grade serous ovarian cancer. Gynecol Oncol 2017;146:170-8.
18. Ji F, Wang Y, Qiu L, Li S, Zhu J, Liang Z, et al. Hypoxia inducible factor 1α-mediated LOX expression correlates with migration and invasion in epithelial ovarian cancer. Int J Oncol 2013;42:1578-88.
19. Payne SL, Fogelgren B, Hess AR, Seftor EA, Wiley EL, Fong SF, et al. Lysyl oxidase regulates breast cancer cell migration and adhesion through a hydrogen peroxide-mediated mechanism. Cancer Res 2005;65:11429-36.
20. Lapointe J, Li C, Higgins JP, Higgins JP, Rijn MV, Bair E, et al. Gene expression profiling identifies clinically relevant subtypes of prostate cancer. Proc Natl Acad Sci U S A 2004;101:811-6.
21. Wang X, Wang Y, Wang GC, Liu PS. miR-29b regulates cell proliferation and invasion in human ovarian clear cell carcinoma by targeting Lysyl oxidase (LOX). Arch Biol Sci 2016;68:155-63.
22. Yang Y, Liang A, Fengling Cui F, Ning LiN, Cong J, Qu l, et al. Lysyl oxidase single-nucleotide polymorphism (SNP) (G473A) is negatively associated with ovarian cancer prognosis. Int J Clin Exp Med 2017;10:16595-602.
23. Schütze F, Röhrig F, Vorlová S, Gätzner S, Kuhn A, Ergün S, et al. Inhibition of lysyl oxidases improves drug diffusion and increases efficacy of cytotoxic treatment in 3D tumor models. Sci Rep 2015;5:17576.
24. Lisowska KM, Olbryt M, Student S, Kujawa KA, Cortez AJ, Simek K, et al. Unsupervised analysis reveals two molecular subgroups of serous ovarian cancer with distinct gene expression profiles and survival. J Cancer Res Clin Oncol 2016;142:1239-52.
25. Zhang Q, Jin XS, Yang ZY, Wei M, Zhu XC, Wang P, et al. Uregulated expression of LOX is a novel independent prognostic marker of worse outcome in gastric cancer patients after curative surgery. Oncol Lett 2013;5:896-902.
26. Wilgus ML, Borczuk AC, Stoopler M, Ginsburg M, Gorenstein L, Sonett JR, et al. Lysyl oxidase: a lung adenocarcinoma biomarker of invasion and survival. Cancer 2011;117:2186-91.
27. Xu X, Wang B, Xu Y. Expression of lysyl oxidase in human osteosarcoma and its clinical significance: a tumor suppressive role of LOX in human osteosarcoma cells. Int J Oncol 2013;43:1578–86.
28. Albinger-Hegyi A, Stoeckli SJ, Schmid S, Storz M, Iotzova G, Probst-Hensch NM, et al. Lysyl oxidase expression is an independent marker of prognosis and a predictor of lymph node metastasis in oral and oropharyngeal squamous cell carcinoma (OSCC). Int J Cancer 2010;126:2653–62.
29. Magnowska M, Zabrowski M, Surowiak P. COX-2 expression pattern is related to ovarian cancer differentiation and prognosis, but is not consistent with new model of pathogenesis; Ginekol Pol 2014;85:335-41.
30. Friedlander ML. Prognostic factors in ovarian cancer. Semin Oncol 1998;25:305–14.
31. Denkert C, Köbel M, Pest S, Koch I, Berger S, Schwabe M, et al. Expression of cyclooxygenase 2 is an independent prognostic factor in human ovarian carcinoma. Am J Pathol 2002;160:893-903.
32. Stadlmann S, Gueth U, Wight E, Kunz-Schughart LA, Hartmann A, Singer G. Expression of peroxisome proliferator activated receptor gamma and cyclo- oxygenase 2 in primary and recurrent ovarian carcinoma. J Clin Pathol 2007;60:307-10.