- Email: [email protected]
Overexpression of COX2 indicates poor survival in urothelial bladder cancer
Annals of Diagnostic Pathology 34 (2018) 50–55
Contents lists available at ScienceDirect
Annals of Diagnostic Pathology journal homepage: www.elsevier.com/locate/anndiagpath
Overexpression of COX2 indicates poor survival in urothelial bladder cancer Usha Agrawal
a,b
a,c
d
e
, Nitu Kumari , Pawan Vasudeva , Nayan K. Mohanty , Sunita Saxena
a,⁎
T
a
National Institute of Pathology, Indian Council of Medical Research, New Delhi-110029, India Faculty of Health and Biomedical Sciences, Symbiosis International University, Lavale, Pune- 412115, India BITS, Pilani, Rajasthan, India d Deptt of Urology, VMMC and Safdarjung Hospital, New Delhi-110029, India e Saket City Hospital, Press Enclave Road, New Delhi -110017 b c
A R T I C L E I N F O
A B S T R A C T
Keywords: COX2 Urothelial bladder cancer
Background: COX2 is a cyclo-oxygenase enzyme expressed in the tumor cells, inflammatory cells, stromal and non-epithelial cells. The study was conducted to evaluate the expression of COX2 in Urothelial carcinoma and find the association with progression and recurrence. Methods: The expression of COX2 was evaluated by real-time PCR and immunohistochemistry. Results: Gene expression of COX2 was found to be upregulated > 28-fold in urothelial cancer compared to adjacent normal bladder mucosa. Inflammatory cell expression of COX2 was found in 92% cases whereas only 37% cases showed COX2 overexpression in tumor cells. Tumor cell COX2 overexpression was significantly associated with invasion and recurrence. Conclusion: COX2 expression is a marker of invasion, recurrence and poor survival and may have a role in predicting the cases which will benefit from additional treatment with COX2 inhibitors in urothelial carcinoma.
1. Introduction Urinary bladder cancer is 7th common cancer globally, in males [1]. Among the various morphological subtypes, urothelial carcinoma accounts for 90% of bladder cancer [2]. The spectrum of urothelial cancer ranges from in situ lesions confined to the mucosa (pTa) to invasive lesions which penetrate the lamina propria (pT1), detrusor muscle (pT2) and extend perivesically (pT3). The morbidity of the disease is increased by frequent recurrences (~70%) in the pTa and pT1 stages [3]. The treatment of pTa and pT1 (Non-muscle invasive urothelial cancer or NMIUC) is conservative surgery with transurethral resection of bladder tumor (TURBT) and in the muscle invasive cancer (MIUC) stages, pT2 and pT3, it is by radical surgery such as cystectomy [4]. Further, to prevent recurrences in non-invasive stages immunomodulators such as BCG are instilled into the bladder directly (intravesical immunotherapy) [5]. Immunotherapy has been found to be of great help in reducing recurrences in about 50% of the cases but the rest recur [6]. At present there is no marker in clinical use to predict the progress or recurrence of disease and while morphologic grade is good at predicting behavior of the tumor it is not a good indicator of
recurrence. It is also difficult to predict which cases will respond to immunotherapy and which will not. However, the response of at least 50% of the tumors to intravesical immunotherapy is an indication of the role the immune response plays in the tumor behavior. A pro-inflammatory immune response is reported to be elicited in most tumors and is protective as the host tissue tends to limit damage by containing the tumor cells [7]. Most commonly reported is the cellmediated immune response where tumor infiltrating lymphocytes (TILs), tumor associated macrophages and NK cells infiltrate into the tumor microenvironment [8]. These immune cells release a multitude of factors which include cytokines, chemokines and growth factors. The factors released have different roles, some promoting tumor cell destruction and some promoting their proliferation. Hence, the end effect on the tumor can only be decided by the way the balance is tipped. One of the pro-inflammatory factors released in tumors is COX2, which is an inducible enzyme belonging to the family of cyclo-oxygenases [9]. COX2 and its enzymatic product PGE2 have a key role in cancer progression [9]. COX2 is encoded by the PTGS2 gene (prostaglandinendoperoxide synthase) and is involved in the conversion of arachidonic acid to prostaglandin H2 which is then converted to
Abbreviations: COX2, cyclo-oxygenase 2; NMIUC, Non-muscle invasive urothelial carcinoma; MIUC, muscle invasive urothelial carcinoma; TURBT, Transurethral resection of bladder tumor; BCG, Bacillus Calmette Guerin; TIL, Tumor infiltrating lymphocytes; NK, Natural Killer; NSAID, Non-steroidal anti-inflammatory drug; GAPDH, Glyceraldehyde 3-phosphate dehydrogenase; RT-PCR, Real-Time Polymerase Chain Reaction; IHC, Immunohistochemistry; mRNA, messenger Ribonucleic acid; FFPE, Formalin fixed paraffin embedded; ANOVA, Analysis of variance ⁎ Corresponding author. E-mail addresses: [email protected] (U. Agrawal), [email protected] (S. Saxena). https://doi.org/10.1016/j.anndiagpath.2018.01.008
1092-9134/ © 2018 Elsevier Inc. All rights reserved.
Annals of Diagnostic Pathology 34 (2018) 50–55
U. Agrawal et al.
prostaglandins (PGD2, PGE2, PGF2α), prostacyclin (PGI2), or thromboxane A2 by tissue-specific isomerases [10, 11]. COX2 expression has been found in the stromal inflammatory compartment as well as in the tumor cells of colorectal carcinoma, cervical carcinoma and others [1215]. The expression of COX2 in tumor cells has been associated with a bad prognosis in cervical cancer and with invasion in breast cancer [14,16,17]. Newer modalities of cancer treatment targeting COX2 are being tried in cancer and include NSAIDs, aspirin and coxibs. There has been initiation of clinical trials testing COX-2 inhibitors for the chemoprevention of a wide variety of cancers such as colorectal, oral, skin, esophageal, and non-small-cell lung cancers and for the treatment of cervical, prostate, and metastatic breast cancers that overexpress COX-2 [18]. In the present study gene expression of COX2 in Urothelial carcinoma was compared to paired normal tissue. The results were validated by studying COX2 protein expression and localization in tumor and stroma by immunohistochemistry. An analysis of association with grade and stage showed increased expression in high grade and muscle invasive cases. Recurrence was found associated with tumor cell expression of COX2.
Table 1 Clinicopathological characteristics of the archival cases. Characteristic
Label
N(%)
Age
< 40 41–60 > 61 Male Female Low grade High grade Non-muscle invasive pTa pT1 Muscle invasive pT2 pT3 Absent Present
28 (8.6%) 151 (46.5%) 146 (44.9%) 271 (83.4%) 54 (16.6%) 159 (48.9%) 166 (51.1%)
Gender Grade Stage
Recurrences
56 (17.2%) 187 (57.5%) 78 (24.0%) 4 (1.2%) 219 (67.4%) 106 (32.6%)
40–60 years. Of the 325 cases, 54 were females and 271 were males. Hematoxylin & eosin stained slides were re-examined to confirm the morphologic diagnosis and included 243 NMIUC (140 low grade, 103 high grade) and 82 MIUC (19 low grade, 63 high grade). The intensity of peritumoral inflammation was estimated independently by two pathologists (UA and SS). The intensity was categorized into four groups (no inflammation = 0, sparse inflammation = 1, moderate inflammation = 2 and dense inflammation with or without formation of lymphoid follicles = 3). FFPE sections (4 μm thick) were taken on poly-L-lysine coated glass slides and processed for immunohistochemistry by non-biotin polymeric technology using Super sensitive one-step polymer-HRP detection kit (Biogenex, Fremont, CA). Heat-induced antigen retrieval was performed with slides placed in citrate buffer (pH 6.0) and slides were incubated with the primary antibody to COX2 (SantaCruz Biotechnology, Dallas, Texas) at 4 °C overnight. DAB (diaminobenzidine) was used as the chromogen. The slides were observed for the expression of COX2 in the inflammatory and tumor cells and the estimated percentage of cells positive for the protein was noted by both pathologists independently and mean of the 2 values was taken for the final analysis. For those cases where the difference in observation between the two pathologists was > 10%, the slides were observed by them together to reach a consensus. The number of cases positive and negative for each component, inflammatory and tumor was analyzed for association with grade and stage. The cases were further categorized as inflammatory only (expression in stromal inflammatory cells only), tumor only (expression in tumor cells alone) and both (expression in inflammatory and tumor cells in the same slide). Each component in the tissue was then scored for COX2 expression in a scale of 0–3 (no expression = 0, < 10% expression = 1, 11–40% expression = 2, and > 40% cells expressing COX2 = 3). COX2 expression in both inflammatory and tumor cells was compared in different grades and stages to analyze the association between them by using the χ2 (Chi-Square) or Fishers' exact test as appropriate. Significance was taken at p < .05.
2. Materials and methods The study included 87 tumor tissue samples and paired normal urinary bladder mucosa collected at the time of surgery in the Department of Urology, Safdarjung Hospital and VMMC, New Delhi, India and also formalin-fixed, paraffin-embedded blocks from 325 cases of urothelial bladder cancer from the archives of the National Institute of Pathology, New Delhi, India. Surgical tissue samples (tumor and paired normal urinary bladder mucosa) was collected separately in 10% buffered formalin and RNAlater. Tissue samples collected in RNAlater were stored at −80 °C till further processing. Formalin-fixed paraffin embedded sections were taken, stained with hematoxylin and eosin and examined for confirmation of diagnosis and morphologic grading and staging according to the 2004 WHO/ISUP criteria [19]. The patients were kept on followup for 36 months and recurrence free survival was noted. The diagnosis on histopathology was 46 NMIUC (29 low grade, 17 high grade, 16 recurrent) and 41 MIUC (9 low grade, 32 high grade) cases. Of the 16 recurrent cases 9 were high grade and 7 were low grade. Samples stored at −80 °C were thawed and 20 mg of tissue was weighed and total RNA was extracted from these samples using the RNeasy mini kit (Qiagen). Genomic DNA contamination was eliminated by DNAse treatment by using RNAse free DNAse kit (Qiagen GmbH, Hilden, Germany). Good quality RNA with readings of > 1.8–2.0 at 260/280 and 260/230 absorbance were utilized for further experiments. Quantitative real-time PCR was performed for detection of gene expression levels of COX2 using TaqMan Universal Master Mix from Applied Biosystems in the ABI Prism 7000 Sequence Detection System. Total RNA was reverse-transcribed by using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystem). Quantitative real-time PCR was performed with 22 ng of cDNA. Each sample was assayed in duplicate in independent reactions. The gene probe for COX2 was FAM labeled (Hs00153133_m1, Applied Biosystem). Target gene expression data were normalized with the expression of housekeeping gene, Glyceraldehyde 3-phosphate dehydrogenase (GAPDH). Relative quantitation was used to calculate Ct values by SDS software (ABI7000) and differential expression between tumor tissue and normal mucosa was identified with the cutoff of p value < .05 (Students t-test) and a mean difference ≥ 2-fold. The gene expression in association with grade and stage was analyzed by Students t-test. For evaluating the immunohistochemical expression of COX2, formalin fixed paraffin embedded (FFPE) blocks of 325 cases of urothelial bladder cancer were retrieved. The demographic details are given in Table 1. The peak age incidence (46.5%) was in the age group
3. Results 3.1. Gene expression of COX2 COX2 mRNA expression was found upregulated in tumor tissue compared to normal mucosa with a mean fold change of > 28-fold. The expression was significantly more in non-muscle invasive tumors as compared to muscle invasive tumors (p = .003). The expression in different pT stages was found significant (p = .029) by ANOVA and multiple comparisons revealed that COX2 expression in pTa and pT1 were each significantly upregulated when compared to pT2 (p = .010 51
Annals of Diagnostic Pathology 34 (2018) 50–55
U. Agrawal et al.
Fig. 1. Boxplots showing COX2 gene expression in various subgroups. a) COX2 expression was not significantly different between low grade and high grade tumors, b) COX2 expression was significantly more in patients with NMIUC compared to those with MIUC, c) COX2 expression was significantly more in both pTa and pT1 compared to pT2 while there is no difference between pTa and pT1, d) Recurrent cases had a higher expression of COX2 but it was not found statistically significant.
(Table 2).
and p = .016 respectively) but there was no difference in expression between pTa and pT1 (p = .751). COX2 expression between low grade and high grade did not show any significant difference (p = .612). Recurrent cases did not show any difference in COX2 expression compared to the expression in non-recurrent cases (p = .387) (Fig. 1). The upregulation of COX2 in tumor was not found associated significantly with survival.
3.3. COX2 in inflammatory cells Immunohistochemical expression of COX2 was observed in the cytoplasm of peritumoral inflammatory cells (Fig. 2a,b) in 304 (93.5%) cases. Expression of COX2 was found to increase with the intensity of inflammation (p < .001). Also, the number of inflammatory cells expressing COX2 was found to be significantly more in high grade cases as compared to low grade cases (p < .001). The number of inflammatory cells expressing COX2 in pT2 was significantly more than that in pTa tumors but no difference in expression was seen between pTa and pT1 (p = .100) or pT1 and pT2 (p = .401). There was no association of COX2 expression in inflammatory cells with recurrence (p = .407).
3.2. Inflammatory cells The intensity of inflammation was found to be sparse in 62 (19.1%) cases, moderate in 132 (40.6%) and dense in 116 (35.7%) cases. The remaining cases did not show inflammation. Intensity of inflammation increased with grade and stage (p < .001) and moderate inflammation was statistically significantly associated with recurrence (p < .001) Table 2 COX2 expression in various grades and stages of urothelial carcinoma of urinary bladder. Characteristic
Age
Gender COX2 in inflammatory cells
COX2 in tumor cells
⁎
Label
< 40 41–60 > 61 Female Male 0 1 2 3 0 1 2 3
n
28 151 146 54 271 21 11 121 172 204 19 81 21
Low grade (n = 159)
14 (8.4) 5 (45.2) 77 (46.4) 24 (15.1) 135 (84.9) 15 (9.4) 6 (3.8) 75 (47.2) 63 (39.6) 95 (59.7) 8 (5.0) 47 (29.6) 9 (5.7)
High grade (n = 166)
14 (8.8) 76 (47.8) 69 (43.4) 30 (18.1) 136 (81.9) 6 (3.6) 5 (3.0) 46 (27.7) 109 (65.7) 109 (65.7) 11 (6.6) 34 (20.5) 12 (7.2)
p < .05.
52
p
0.863
0.471 < 0.001⁎
0.284
Non-invasive
Muscle-invasive
pTa (56)
pT1 (187)
pT2 + pT3 (82)
7 (12.5) 29 (51.8) 20 (35.7) 10 (17.9) 46 (82.1) 3 (5.4) 2 (3.6) 31 (55.4) 20 (35.7) 27 (48.2) 3 (5.4) 21 (37.5) 5 (8.9)
14 (7.5) 87 (46.5) 86 (46.0) 33 (17.6) 154 (82.4) 12 (6.4) 5 (2.7) 63 (33.7) 107 (57.2) 125 (66.8) 9 (4.8) 44 (23.5) 9 (4.8)
7 (8.5) 35 (42.7) 40 (48.8) 11 (13.4) 71 (86.6) 6 (7.3) 4 (4.9) 27 (32.9) 45 (54.9) 52 (63.4) 7 (8.5) 16 (19.5) 7 (8.5)
P
0.520
0.666 0.115
0.002⁎
Annals of Diagnostic Pathology 34 (2018) 50–55
U. Agrawal et al.
Fig. 2. COX2 expression by immunohistochemistry shows a) inflammatory cells expressing COX2 in the cytoplasm, 100× b) moderate inflammation with occasional inflammatory cell expressing COX2 as denoted by the arrows, 200× c) Tumor cells of Urothelial muscle invasive carcinoma expressing cytoplasmic COX2, d) Both inflammatory cells and tumor cells expressing COX2 in a case of nonmuscle invasive Urothelial carcinoma.
COX3 which represents a splice variant of COX1 that encodes a truncated protein lacking enzymatic activity [9]. Among other stimuli, COX2 is induced by cytokines which are released in the tumor microenvironment because of the immune cell infiltration. The overexpression of COX2 has been reported in various cell types including malignant cells, stromal cells, epithelial cells, and nonepithelial cells [20]. The role of the COX2/PGE2 pathway has been implicated in each of the hallmarks of cancer which include sustained proliferative signaling, evading growth suppressors, resisting cell death, enabling replicative immortality, inducing angiogenesis, activating invasion and metastasis, reprogramming of energy metabolism and evading immune destruction [9]. Increased COX2 expression promotes prostaglandin production which promotes proliferation, angiogenesis and ultimately metastasis by activation of tyrosine kinase receptors. COX2 also stimulates production of VEGF and bFGF which are both known as pro-angiogenic factors. It also stimulates anti-apoptotic pathways and prevents endothelial cell apoptosis thus promoting angiogenesis [21]. COX2 is found to have a role in tumor growth and is reported to be upregulated in most cancers including bladder, breast, pancreatic, and ovarian cancers [16, 22-24]. Aberrant expression of COX2 in tumor cells and association with invasion and metastasis has also been reported in breast cancer. COX2 overexpression has been demonstrated to increase motility of breast cancer cells through a matrigel membrane and increase the cell migration and cell invasion > 2-fold while addition of an inhibitor decreased invasion by 54%. The increased mobility has been attributed to cytoskeletal alterations and basement membrane degradation due to increased expression of pro-urokinase plasminogen activator (pro-uPA) [25]. High COX2 expression is also a predictor of biochemical recurrence in prostate cancer [26]. In colon, COX2 was found to be overexpressed in 50% of benign polyps and 80–85% of adenocarcinomas [27]. High COX2 expression by immunohistochemistry has also been found to correlate with colon cancer recurrence and hematogenous metastasis [28]. COX2 protein expression by IHC was also found to be more in the bladder cancer cases with recurrence than in cases without recurrence, and though the sample size in each group was only 5 and 6 respectively it suggests that
3.4. COX2 in tumor cells COX2 expression was found in the cytoplasm of tumor cells (Fig. 2c,d) in 121 (37.2%) cases but the expression was not found related to grade (p = .153). Of the 121 cases with COX2 expression in tumor cells most were of pT1 stage (51.2%) but higher number of cells (> 50%) were positive in pT2 stage (p = .011). A qualitative analysis of COX2 positive/negative expression showed that 51.9% cases with COX2 expression in tumor cells showed recurrence as opposed to 30.1% with COX2 negative profile (p < .001) (Table 3). COX2 expression in larger percentage of tumor cells was associated with poor survival (p = .001). However with increasing time the cumulative survival diminishes even with negative COX2 expression (Fig. 3) suggesting that COX2 is a prognostic marker in the early months after diagnosis. 4. Discussion The cyclo-oxygenase enzymes catalyze a key step in arachidonic acid metabolism. There are 3 COX isoforms, COX1 which is constitutively expressed in most tissues, COX2 which is induced momentarily and strongly in response to many physiological stimuli and Table 3 COX2 expression (qualitative) in inflammatory and tumor cells. Characteristic
Grade
Stage
Recurrence
⁎
Low High p NMIUC MIUC p No Yes p
COX2 in inflammatory cells
COX2 in tumor cells
Negative
Positive
Negative
Positive
15 (9.4%) 6 (3.6%) 0.042⁎ 15 (6.2%) 6 (7.3%) 0.716 16 (7.3%) 5 (4.7%) 0.474
144 (90.6%) 160 (96.4%)
95 (59.7%) 109 (65.7%) 0.270 152 (62.6%) 52 (63.4%) 0.889 153 (69.9%) 51 (48.1%) < 0.001⁎
64 (40.3%) 57 (34.3%)
228 (93.8%) 76 (92.7%) 203 (92.7%) 101 (95.3%)
91 (37.4%) 30 (36.6%) 66 (30.1%) 55 (51.9%)
p < .05.
53
Annals of Diagnostic Pathology 34 (2018) 50–55
U. Agrawal et al.
Fig. 3. Kaplan Meier survival plot shows that patient survival is best when COX2 expression is negative (66.8% cumulative survival at 41 months of follow-up) but with increase in time the survival is almost equal irrespective of COX2 status, falling to 31.0% at the 48th month in COX2 negative individuals. COX2 positive in > 41% tumor cells indicates a bad prognosis with a cumulative survival of 12.1% at 35 months follow-up.
colorectal cancer risk by suppressing the immune response. NSAIDs block endogenous prostaglandin synthesis through inhibition of cyclooxygenase enzymatic activity, both COX1 and COX2 [34]. Similar decrease in bladder cancer risk has been reported with the use of NSAIDs [29]. COX2 specific inhibitors, the coxibs are also now available and act by down-regulating bcl2 (which is a suppressor of apoptosis) [35, 36]. Inhibition of COX2 causes reduction of bladder cancer chemoresistance and increases apoptosis of tumor including cancer stem cells [36].
COX-2 contributes to superficial bladder cancer recurrence [29]. COX2 expression has also been reported to be higher in muscle-invasive bladder cancer in a meta-analysis conducted by Czachorowski et al. However, the same group reported only a weak association with recurrence [22]. The distribution of cases in our centre shows a low incidence of PUNLMP (papillary urothelial neoplasia of unknown malignant potential) cases and of low grade muscle invasive urothelial carcinoma cases. While Tokgoz et al., 2007 [30] report almost 50% low grade tumors in the muscle invasive urothelial carcinomas, Jimenez et al., 2000 [31] have reported very few low grade cases in their cohort. The variation in incidence in different cohorts could also be attributed to the availability of access to advanced health care facility. PUNLMP case may recur and progress. Maxwell et al., 2015 reported a progression of PUNLMP to low grade urothelial carcinoma in 9.5% patients and to high grade urothelial carcinoma in 1.6% patients [32]. In the same year, Zhang et al. reported a recurrence in 19.7% and progression up to pTa in 16.9%. They reported that increased mitosis, tumor multifocality and age at diagnosis were independent prognostic predictors [33]. However, the role of biomarkers in predicting prognosis has not been well studied. However, localization of COX2 in either inflammatory, stromal or malignant cell component and the association of overexpression of COX2 in any one of the component with grade, stage and recurrence has not been explored thoroughly. In the present study the upregulation of the gene and expression of COX2 in inflammatory cells was not found to be associated with stage of the tumor. However, inflammatory cell expression of COX2 was found to be associated with high grade urothelial cancer, possibly reflecting host immune response to high grade tumors. Tumor cell expression of COX2 has been found to be associated with muscle invasion (p = .011) and recurrence (p < .001) in our study. The role of COX2 could not be evaluated in cases of PUNLMP as we had only 2 cases on record. Further evaluation of COX2 in such cases could help to identify its role as a prognostic indicator in early stages. Inhibition of COX2 in aspirin users has been found to decrease
5. Conclusion The association of COX2 with progression of disease, invasion, metastasis and recurrence along with the epidemiological evidence of decreased risk in habitual users of NSAIDs suggests a strong role of COX2 inhibitors in disease management. The present study shows the specific association of tumor cells expressing COX2 with recurrence and invasion of urothelial cancers indicating its role as a marker of aggressive behavior and also a beneficial role for targeting of this enzyme in management of Urothelial cancer patients. Acknowledgements The authors would like to acknowledge the Indian Council of Medical Research (53/4/2008-BMS) for financial support. References [1] Ploeg M, Aben KKH, Kiemeney LA. The present and future burden of urinary bladder cancer in the world. World J Urol 2009;27(3):289–93. [2] Heney NM. Natural history of superficial bladder cancer. Prognostic features and long-term disease course. Urol Clin North Am 1992;19:429–33. [3] Pasin E, Josephson DY, Mitra AP, Cote RJ, Stein JP. Superficial bladder cancer: an update on etiology, molecular development, classification, and natural history. Rev Urol 2008;10(1):31–43. [4] McDougal WS, Shipley WU, Kaufman DS, Dahl DM, Michaelson MD, Zietman AL. Cancer of the bladder, ureter and renal pelvis. In: DeVitaHellmanRosenbergeditors.
54
Annals of Diagnostic Pathology 34 (2018) 50–55
U. Agrawal et al.
[5]
[6]
[7] [8] [9]
[10]
[11] [12]
[13]
[14]
[15]
[16]
[17] [18] [19]
[20]
Neurochem 2000;75:2020–8. [21] Kuwano T, Nakao S, Yamamoto H, Tsuneyoshi M, Yamamoto T, Kuwano M, et al. Cyclooxygenase 2 is a key enzyme for inflammatory cytokine-induced angiogenesis. FASEB J 2004;18:300–10. [22] Czachorowski MJ, AFS Amaral, Montes-Moreno S, Lloreta J, Carrato A, Tardón A, et al. Cyclooxygenase-2 expression in bladder cancer and patient prognosis: results from a large clinical cohort and meta-analysis. PLoS ONE 2012;7(9):e45025. [23] Hill R, Li Y, Tran LM, Dry S, Calvopina JH, Garcia A, et al. Cell intrinsic role of COX2 in pancreatic cancer development. Mol Cancer Ther 2012;11(10):2127–37. [24] Ferrandina G, Lauriola L, Zannoni GF, Fagotti A, Fanfani F, Legge F, et al. Increased cyclooxygenase-2 (COX-2) expression is associated with chemotherapy resistance and outcome in ovarian cancer patients. Ann Oncol 2002;13(8):1205–11. [25] Singh B, Berry JA, Shoher A, Ramakrishnan V, Lucci A. COX-2 overexpression increases motility and invasion of breast cancer cells. Int J Oncol 2005;26(5):1393–9. [26] Cohen BL, Gomez P, Omori Y, Duncan RC, Civantos F, Soloway MS, et al. Cyclooxygenase-2 (COX-2) expression is an independent predictor of prostate cancer recurrence. Int J Cancer 2006;1;119(5):1082–7. [27] Williams CS, Mann M, RN DuBois. The role of cyclooxygenases in inflammation, cancer, and development. Oncogene 1999;18:7908–16. [28] Tomozawa S, Tsuno NH, Sunami E, Hatano K, Kitayama J, Osada T, et al. Cyclooxygenase-2 overexpression correlates with tumour recurrence, especially haematogenous metastasis, of colorectal cancer. Br J Cancer 2000;83(3):324–8. [29] Okajima E, Uemura H, Ohnishi S, Tanaka M, Ohta M, Tani M, et al. Expression of cyclooxygenase-2 in primary superficial bladder cancer tissue may predict risk of its recurrence after complete transurethral resection. Aktuelle Urol 2003;34(4):256–8. [30] Tokgoz Husnu, Turkolmez Kadir, Resorlu Berkan, Kose Kenan, Tulunay Ozden, Beduk Yasar. Pathological staging of muscle invasive bladder cancer: is substaging of pT2 tumors really necessary? Int Braz J Urol 2007;33(6):777–84. [31] Jimenez RE, Gheiler E, Oskanian P, Tiguert R, Sakr W, Wood Jr. DP, et al. Grading the invasive component of urothelial carcinoma of the bladder and its relationship with progression free survival. Am J Surg Pathol 2000;24(7):980–7. [32] Maxwell JP, Wang C, Wiebe N, Yilmaz A, Trpkov K. Long-term outcome of primary papillary urothelial neoplasm of low malignant potential (PUNLMP) including PUNLMP with inverted growth. Diagn Pathol 2015;10(3). [33] Zhang XK, Wang YY, Chen JW, Qin T. Bladder papillary urothelial neoplasm of low malignant potential in Chinese: a clinical and pathological analysis. Int J Clin Exp Pathol 2015;8(5):5549–55. [34] Xu XC. COX-2 inhibitors in cancer treatment and prevention, a recent development. Anticancer Drugs 2002;13(2):127–37. [35] Steinbach G, Lynch PM, Phillips RK, Wallace MH, Hawk E, Gordon GB, et al. The effect of celecoxib, a cyclooxygenase-2 inhibitor, in familial adenomatous polyposis. N. Engl. J. Med. 2000;342:1946–52. [36] Kurtova AV, Xiao J, Mo Q, Pazhanisamy S, Krasnow R, Lerner SP, et al. Blocking PGE2-induced tumour repopulation abrogates bladder cancer chemoresistance. Nature 2015;517:209–13.
Cancer, Principles and Practice of Oncology. 9th editionPhiladelphia: Lippincott Williams & Wilkins; 2011. p. 1192–211. Zlotta AR, van Vooren JP, Huygen K, et al. What is the optimal regimen for BCG intravesical therapy? Are six weekly instillations necessary? Eur Urol 2000;37(4):470–7. Lebret T, Bohin D, Kassardjian Z, Herve JM, Molinie V, Barre P, et al. Recurrence, progression and success in stage Ta grade 3 bladder tumors treated with low dose bacillus Calmette-Guerin instillations. J Urol 2000;163(1):63–7. Grivennikov SI, Greten FR, Karin M. Immunity, inflammation, and cancer. Cell 2010;140(6):883–99. Gajewski TF, Schreiber H, Fu Y. Innate and adaptive immune cells in the tumor microenvironment. Nat Immunol 2013;14:1014–22. Greenhough A, Smartt HJM, Moore AE, Roberts HR, Williams AC, Paraskeva C, et al. The COX-2/PGE2 pathway: key roles in the hallmarks of cancer and adaptation to the tumour microenvironment. Carcinogenesis 2009;30(3):377–86. Vogel LK, Sæbø M, Høyer H, Kopp TI, Vogel U, Godiksen S, et al. Intestinal PTGS2 mRNA levels, PTGS2 gene polymorphisms, and colorectal carcinogenesis. PLoS ONE 2014;9(8):e105254. Allaj V, Guo C, Nie D. Non-steroid anti-inflammatory drugs, prostaglandins, and cancer. Cell Biosci 2013;3:8. Asting AG, Carén H, Andersson M, Lönnroth C, Lagerstedt K, Lundholm K. COX-2 gene expression in colon cancer tissue related to regulating factors and promoter methylation status. BMC Cancer 2011;11(238). Roelofs HMJ, te Morsche RHM, van Heumen BWH, Nagengast FM, Peters WHM. Over-expression of COX-2 mRNA in colorectal cancer. BMC Gastroenterol 2014;14(1). Ferrandina G, Lauriola L, Zannoni GF, et al. Expression of cyclooxygenase-2 (COX2) in tumour and stroma compartments in cervical cancer: clinical implications. Br J Cancer 2002;87(10):1145–52. Ferrandina G, Ranelletti FO, Legge F, Gessi M, Salutari V, Distefano MG, et al. Prognostic role of the ratio between Cyclooxygenase-2 in tumor and stroma compartments in cervical cancer. Clin Cancer Res 2004;10:3117–23. Kim M-J, Kim H-S, Lee S-H, Yang Y, Lee M-S, Lim J-S. NDRG2 controls COX-2/ PGE2-mediated breast cancer cell migration and invasion. Mol Cells 2014;37(10):759–65. Jana D, Sarkar DK, Ganguly S, Saha S, Sa G, Manna AK, et al. Role of cyclooxygenase 2 (COX-2) in prognosis of breast cancer. Ind J Surg Oncol 2014;5(1):59–65. Michalowski J. COX-2 inhibitors: cancer trials test new uses for pain drug. JNCI J Natl Cancer Inst 2002;94(4):248–9. Sauter G, Algaba F, Amin MB, et al. Non-invasive urothelial tumours. In: Eble JN, Sauter G, Epstein JI, Sesterhenn IA, editors. World Health Organization Classification of Tumours: Pathology and Genetics of Tumours of the Urinary System and Male Genital Organ. Lyon, France: IARC Press; 2004. p. 110–23. Fiebich BL, Mueksch B, Boehringer M, Hull M. Interleukin-1beta induces cyclooxygenase-2 and prostaglandin E(2) synthesis in human neuroblastoma cells: involvement of p38 mitogen-activated protein kinase and nuclear factorkappaB. J
55
Contents lists available at ScienceDirect
Annals of Diagnostic Pathology journal homepage: www.elsevier.com/locate/anndiagpath
Overexpression of COX2 indicates poor survival in urothelial bladder cancer Usha Agrawal
a,b
a,c
d
e
, Nitu Kumari , Pawan Vasudeva , Nayan K. Mohanty , Sunita Saxena
a,⁎
T
a
National Institute of Pathology, Indian Council of Medical Research, New Delhi-110029, India Faculty of Health and Biomedical Sciences, Symbiosis International University, Lavale, Pune- 412115, India BITS, Pilani, Rajasthan, India d Deptt of Urology, VMMC and Safdarjung Hospital, New Delhi-110029, India e Saket City Hospital, Press Enclave Road, New Delhi -110017 b c
A R T I C L E I N F O
A B S T R A C T
Keywords: COX2 Urothelial bladder cancer
Background: COX2 is a cyclo-oxygenase enzyme expressed in the tumor cells, inflammatory cells, stromal and non-epithelial cells. The study was conducted to evaluate the expression of COX2 in Urothelial carcinoma and find the association with progression and recurrence. Methods: The expression of COX2 was evaluated by real-time PCR and immunohistochemistry. Results: Gene expression of COX2 was found to be upregulated > 28-fold in urothelial cancer compared to adjacent normal bladder mucosa. Inflammatory cell expression of COX2 was found in 92% cases whereas only 37% cases showed COX2 overexpression in tumor cells. Tumor cell COX2 overexpression was significantly associated with invasion and recurrence. Conclusion: COX2 expression is a marker of invasion, recurrence and poor survival and may have a role in predicting the cases which will benefit from additional treatment with COX2 inhibitors in urothelial carcinoma.
1. Introduction Urinary bladder cancer is 7th common cancer globally, in males [1]. Among the various morphological subtypes, urothelial carcinoma accounts for 90% of bladder cancer [2]. The spectrum of urothelial cancer ranges from in situ lesions confined to the mucosa (pTa) to invasive lesions which penetrate the lamina propria (pT1), detrusor muscle (pT2) and extend perivesically (pT3). The morbidity of the disease is increased by frequent recurrences (~70%) in the pTa and pT1 stages [3]. The treatment of pTa and pT1 (Non-muscle invasive urothelial cancer or NMIUC) is conservative surgery with transurethral resection of bladder tumor (TURBT) and in the muscle invasive cancer (MIUC) stages, pT2 and pT3, it is by radical surgery such as cystectomy [4]. Further, to prevent recurrences in non-invasive stages immunomodulators such as BCG are instilled into the bladder directly (intravesical immunotherapy) [5]. Immunotherapy has been found to be of great help in reducing recurrences in about 50% of the cases but the rest recur [6]. At present there is no marker in clinical use to predict the progress or recurrence of disease and while morphologic grade is good at predicting behavior of the tumor it is not a good indicator of
recurrence. It is also difficult to predict which cases will respond to immunotherapy and which will not. However, the response of at least 50% of the tumors to intravesical immunotherapy is an indication of the role the immune response plays in the tumor behavior. A pro-inflammatory immune response is reported to be elicited in most tumors and is protective as the host tissue tends to limit damage by containing the tumor cells [7]. Most commonly reported is the cellmediated immune response where tumor infiltrating lymphocytes (TILs), tumor associated macrophages and NK cells infiltrate into the tumor microenvironment [8]. These immune cells release a multitude of factors which include cytokines, chemokines and growth factors. The factors released have different roles, some promoting tumor cell destruction and some promoting their proliferation. Hence, the end effect on the tumor can only be decided by the way the balance is tipped. One of the pro-inflammatory factors released in tumors is COX2, which is an inducible enzyme belonging to the family of cyclo-oxygenases [9]. COX2 and its enzymatic product PGE2 have a key role in cancer progression [9]. COX2 is encoded by the PTGS2 gene (prostaglandinendoperoxide synthase) and is involved in the conversion of arachidonic acid to prostaglandin H2 which is then converted to
Abbreviations: COX2, cyclo-oxygenase 2; NMIUC, Non-muscle invasive urothelial carcinoma; MIUC, muscle invasive urothelial carcinoma; TURBT, Transurethral resection of bladder tumor; BCG, Bacillus Calmette Guerin; TIL, Tumor infiltrating lymphocytes; NK, Natural Killer; NSAID, Non-steroidal anti-inflammatory drug; GAPDH, Glyceraldehyde 3-phosphate dehydrogenase; RT-PCR, Real-Time Polymerase Chain Reaction; IHC, Immunohistochemistry; mRNA, messenger Ribonucleic acid; FFPE, Formalin fixed paraffin embedded; ANOVA, Analysis of variance ⁎ Corresponding author. E-mail addresses: [email protected] (U. Agrawal), [email protected] (S. Saxena). https://doi.org/10.1016/j.anndiagpath.2018.01.008
1092-9134/ © 2018 Elsevier Inc. All rights reserved.
Annals of Diagnostic Pathology 34 (2018) 50–55
U. Agrawal et al.
prostaglandins (PGD2, PGE2, PGF2α), prostacyclin (PGI2), or thromboxane A2 by tissue-specific isomerases [10, 11]. COX2 expression has been found in the stromal inflammatory compartment as well as in the tumor cells of colorectal carcinoma, cervical carcinoma and others [1215]. The expression of COX2 in tumor cells has been associated with a bad prognosis in cervical cancer and with invasion in breast cancer [14,16,17]. Newer modalities of cancer treatment targeting COX2 are being tried in cancer and include NSAIDs, aspirin and coxibs. There has been initiation of clinical trials testing COX-2 inhibitors for the chemoprevention of a wide variety of cancers such as colorectal, oral, skin, esophageal, and non-small-cell lung cancers and for the treatment of cervical, prostate, and metastatic breast cancers that overexpress COX-2 [18]. In the present study gene expression of COX2 in Urothelial carcinoma was compared to paired normal tissue. The results were validated by studying COX2 protein expression and localization in tumor and stroma by immunohistochemistry. An analysis of association with grade and stage showed increased expression in high grade and muscle invasive cases. Recurrence was found associated with tumor cell expression of COX2.
Table 1 Clinicopathological characteristics of the archival cases. Characteristic
Label
N(%)
Age
< 40 41–60 > 61 Male Female Low grade High grade Non-muscle invasive pTa pT1 Muscle invasive pT2 pT3 Absent Present
28 (8.6%) 151 (46.5%) 146 (44.9%) 271 (83.4%) 54 (16.6%) 159 (48.9%) 166 (51.1%)
Gender Grade Stage
Recurrences
56 (17.2%) 187 (57.5%) 78 (24.0%) 4 (1.2%) 219 (67.4%) 106 (32.6%)
40–60 years. Of the 325 cases, 54 were females and 271 were males. Hematoxylin & eosin stained slides were re-examined to confirm the morphologic diagnosis and included 243 NMIUC (140 low grade, 103 high grade) and 82 MIUC (19 low grade, 63 high grade). The intensity of peritumoral inflammation was estimated independently by two pathologists (UA and SS). The intensity was categorized into four groups (no inflammation = 0, sparse inflammation = 1, moderate inflammation = 2 and dense inflammation with or without formation of lymphoid follicles = 3). FFPE sections (4 μm thick) were taken on poly-L-lysine coated glass slides and processed for immunohistochemistry by non-biotin polymeric technology using Super sensitive one-step polymer-HRP detection kit (Biogenex, Fremont, CA). Heat-induced antigen retrieval was performed with slides placed in citrate buffer (pH 6.0) and slides were incubated with the primary antibody to COX2 (SantaCruz Biotechnology, Dallas, Texas) at 4 °C overnight. DAB (diaminobenzidine) was used as the chromogen. The slides were observed for the expression of COX2 in the inflammatory and tumor cells and the estimated percentage of cells positive for the protein was noted by both pathologists independently and mean of the 2 values was taken for the final analysis. For those cases where the difference in observation between the two pathologists was > 10%, the slides were observed by them together to reach a consensus. The number of cases positive and negative for each component, inflammatory and tumor was analyzed for association with grade and stage. The cases were further categorized as inflammatory only (expression in stromal inflammatory cells only), tumor only (expression in tumor cells alone) and both (expression in inflammatory and tumor cells in the same slide). Each component in the tissue was then scored for COX2 expression in a scale of 0–3 (no expression = 0, < 10% expression = 1, 11–40% expression = 2, and > 40% cells expressing COX2 = 3). COX2 expression in both inflammatory and tumor cells was compared in different grades and stages to analyze the association between them by using the χ2 (Chi-Square) or Fishers' exact test as appropriate. Significance was taken at p < .05.
2. Materials and methods The study included 87 tumor tissue samples and paired normal urinary bladder mucosa collected at the time of surgery in the Department of Urology, Safdarjung Hospital and VMMC, New Delhi, India and also formalin-fixed, paraffin-embedded blocks from 325 cases of urothelial bladder cancer from the archives of the National Institute of Pathology, New Delhi, India. Surgical tissue samples (tumor and paired normal urinary bladder mucosa) was collected separately in 10% buffered formalin and RNAlater. Tissue samples collected in RNAlater were stored at −80 °C till further processing. Formalin-fixed paraffin embedded sections were taken, stained with hematoxylin and eosin and examined for confirmation of diagnosis and morphologic grading and staging according to the 2004 WHO/ISUP criteria [19]. The patients were kept on followup for 36 months and recurrence free survival was noted. The diagnosis on histopathology was 46 NMIUC (29 low grade, 17 high grade, 16 recurrent) and 41 MIUC (9 low grade, 32 high grade) cases. Of the 16 recurrent cases 9 were high grade and 7 were low grade. Samples stored at −80 °C were thawed and 20 mg of tissue was weighed and total RNA was extracted from these samples using the RNeasy mini kit (Qiagen). Genomic DNA contamination was eliminated by DNAse treatment by using RNAse free DNAse kit (Qiagen GmbH, Hilden, Germany). Good quality RNA with readings of > 1.8–2.0 at 260/280 and 260/230 absorbance were utilized for further experiments. Quantitative real-time PCR was performed for detection of gene expression levels of COX2 using TaqMan Universal Master Mix from Applied Biosystems in the ABI Prism 7000 Sequence Detection System. Total RNA was reverse-transcribed by using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystem). Quantitative real-time PCR was performed with 22 ng of cDNA. Each sample was assayed in duplicate in independent reactions. The gene probe for COX2 was FAM labeled (Hs00153133_m1, Applied Biosystem). Target gene expression data were normalized with the expression of housekeeping gene, Glyceraldehyde 3-phosphate dehydrogenase (GAPDH). Relative quantitation was used to calculate Ct values by SDS software (ABI7000) and differential expression between tumor tissue and normal mucosa was identified with the cutoff of p value < .05 (Students t-test) and a mean difference ≥ 2-fold. The gene expression in association with grade and stage was analyzed by Students t-test. For evaluating the immunohistochemical expression of COX2, formalin fixed paraffin embedded (FFPE) blocks of 325 cases of urothelial bladder cancer were retrieved. The demographic details are given in Table 1. The peak age incidence (46.5%) was in the age group
3. Results 3.1. Gene expression of COX2 COX2 mRNA expression was found upregulated in tumor tissue compared to normal mucosa with a mean fold change of > 28-fold. The expression was significantly more in non-muscle invasive tumors as compared to muscle invasive tumors (p = .003). The expression in different pT stages was found significant (p = .029) by ANOVA and multiple comparisons revealed that COX2 expression in pTa and pT1 were each significantly upregulated when compared to pT2 (p = .010 51
Annals of Diagnostic Pathology 34 (2018) 50–55
U. Agrawal et al.
Fig. 1. Boxplots showing COX2 gene expression in various subgroups. a) COX2 expression was not significantly different between low grade and high grade tumors, b) COX2 expression was significantly more in patients with NMIUC compared to those with MIUC, c) COX2 expression was significantly more in both pTa and pT1 compared to pT2 while there is no difference between pTa and pT1, d) Recurrent cases had a higher expression of COX2 but it was not found statistically significant.
(Table 2).
and p = .016 respectively) but there was no difference in expression between pTa and pT1 (p = .751). COX2 expression between low grade and high grade did not show any significant difference (p = .612). Recurrent cases did not show any difference in COX2 expression compared to the expression in non-recurrent cases (p = .387) (Fig. 1). The upregulation of COX2 in tumor was not found associated significantly with survival.
3.3. COX2 in inflammatory cells Immunohistochemical expression of COX2 was observed in the cytoplasm of peritumoral inflammatory cells (Fig. 2a,b) in 304 (93.5%) cases. Expression of COX2 was found to increase with the intensity of inflammation (p < .001). Also, the number of inflammatory cells expressing COX2 was found to be significantly more in high grade cases as compared to low grade cases (p < .001). The number of inflammatory cells expressing COX2 in pT2 was significantly more than that in pTa tumors but no difference in expression was seen between pTa and pT1 (p = .100) or pT1 and pT2 (p = .401). There was no association of COX2 expression in inflammatory cells with recurrence (p = .407).
3.2. Inflammatory cells The intensity of inflammation was found to be sparse in 62 (19.1%) cases, moderate in 132 (40.6%) and dense in 116 (35.7%) cases. The remaining cases did not show inflammation. Intensity of inflammation increased with grade and stage (p < .001) and moderate inflammation was statistically significantly associated with recurrence (p < .001) Table 2 COX2 expression in various grades and stages of urothelial carcinoma of urinary bladder. Characteristic
Age
Gender COX2 in inflammatory cells
COX2 in tumor cells
⁎
Label
< 40 41–60 > 61 Female Male 0 1 2 3 0 1 2 3
n
28 151 146 54 271 21 11 121 172 204 19 81 21
Low grade (n = 159)
14 (8.4) 5 (45.2) 77 (46.4) 24 (15.1) 135 (84.9) 15 (9.4) 6 (3.8) 75 (47.2) 63 (39.6) 95 (59.7) 8 (5.0) 47 (29.6) 9 (5.7)
High grade (n = 166)
14 (8.8) 76 (47.8) 69 (43.4) 30 (18.1) 136 (81.9) 6 (3.6) 5 (3.0) 46 (27.7) 109 (65.7) 109 (65.7) 11 (6.6) 34 (20.5) 12 (7.2)
p < .05.
52
p
0.863
0.471 < 0.001⁎
0.284
Non-invasive
Muscle-invasive
pTa (56)
pT1 (187)
pT2 + pT3 (82)
7 (12.5) 29 (51.8) 20 (35.7) 10 (17.9) 46 (82.1) 3 (5.4) 2 (3.6) 31 (55.4) 20 (35.7) 27 (48.2) 3 (5.4) 21 (37.5) 5 (8.9)
14 (7.5) 87 (46.5) 86 (46.0) 33 (17.6) 154 (82.4) 12 (6.4) 5 (2.7) 63 (33.7) 107 (57.2) 125 (66.8) 9 (4.8) 44 (23.5) 9 (4.8)
7 (8.5) 35 (42.7) 40 (48.8) 11 (13.4) 71 (86.6) 6 (7.3) 4 (4.9) 27 (32.9) 45 (54.9) 52 (63.4) 7 (8.5) 16 (19.5) 7 (8.5)
P
0.520
0.666 0.115
0.002⁎
Annals of Diagnostic Pathology 34 (2018) 50–55
U. Agrawal et al.
Fig. 2. COX2 expression by immunohistochemistry shows a) inflammatory cells expressing COX2 in the cytoplasm, 100× b) moderate inflammation with occasional inflammatory cell expressing COX2 as denoted by the arrows, 200× c) Tumor cells of Urothelial muscle invasive carcinoma expressing cytoplasmic COX2, d) Both inflammatory cells and tumor cells expressing COX2 in a case of nonmuscle invasive Urothelial carcinoma.
COX3 which represents a splice variant of COX1 that encodes a truncated protein lacking enzymatic activity [9]. Among other stimuli, COX2 is induced by cytokines which are released in the tumor microenvironment because of the immune cell infiltration. The overexpression of COX2 has been reported in various cell types including malignant cells, stromal cells, epithelial cells, and nonepithelial cells [20]. The role of the COX2/PGE2 pathway has been implicated in each of the hallmarks of cancer which include sustained proliferative signaling, evading growth suppressors, resisting cell death, enabling replicative immortality, inducing angiogenesis, activating invasion and metastasis, reprogramming of energy metabolism and evading immune destruction [9]. Increased COX2 expression promotes prostaglandin production which promotes proliferation, angiogenesis and ultimately metastasis by activation of tyrosine kinase receptors. COX2 also stimulates production of VEGF and bFGF which are both known as pro-angiogenic factors. It also stimulates anti-apoptotic pathways and prevents endothelial cell apoptosis thus promoting angiogenesis [21]. COX2 is found to have a role in tumor growth and is reported to be upregulated in most cancers including bladder, breast, pancreatic, and ovarian cancers [16, 22-24]. Aberrant expression of COX2 in tumor cells and association with invasion and metastasis has also been reported in breast cancer. COX2 overexpression has been demonstrated to increase motility of breast cancer cells through a matrigel membrane and increase the cell migration and cell invasion > 2-fold while addition of an inhibitor decreased invasion by 54%. The increased mobility has been attributed to cytoskeletal alterations and basement membrane degradation due to increased expression of pro-urokinase plasminogen activator (pro-uPA) [25]. High COX2 expression is also a predictor of biochemical recurrence in prostate cancer [26]. In colon, COX2 was found to be overexpressed in 50% of benign polyps and 80–85% of adenocarcinomas [27]. High COX2 expression by immunohistochemistry has also been found to correlate with colon cancer recurrence and hematogenous metastasis [28]. COX2 protein expression by IHC was also found to be more in the bladder cancer cases with recurrence than in cases without recurrence, and though the sample size in each group was only 5 and 6 respectively it suggests that
3.4. COX2 in tumor cells COX2 expression was found in the cytoplasm of tumor cells (Fig. 2c,d) in 121 (37.2%) cases but the expression was not found related to grade (p = .153). Of the 121 cases with COX2 expression in tumor cells most were of pT1 stage (51.2%) but higher number of cells (> 50%) were positive in pT2 stage (p = .011). A qualitative analysis of COX2 positive/negative expression showed that 51.9% cases with COX2 expression in tumor cells showed recurrence as opposed to 30.1% with COX2 negative profile (p < .001) (Table 3). COX2 expression in larger percentage of tumor cells was associated with poor survival (p = .001). However with increasing time the cumulative survival diminishes even with negative COX2 expression (Fig. 3) suggesting that COX2 is a prognostic marker in the early months after diagnosis. 4. Discussion The cyclo-oxygenase enzymes catalyze a key step in arachidonic acid metabolism. There are 3 COX isoforms, COX1 which is constitutively expressed in most tissues, COX2 which is induced momentarily and strongly in response to many physiological stimuli and Table 3 COX2 expression (qualitative) in inflammatory and tumor cells. Characteristic
Grade
Stage
Recurrence
⁎
Low High p NMIUC MIUC p No Yes p
COX2 in inflammatory cells
COX2 in tumor cells
Negative
Positive
Negative
Positive
15 (9.4%) 6 (3.6%) 0.042⁎ 15 (6.2%) 6 (7.3%) 0.716 16 (7.3%) 5 (4.7%) 0.474
144 (90.6%) 160 (96.4%)
95 (59.7%) 109 (65.7%) 0.270 152 (62.6%) 52 (63.4%) 0.889 153 (69.9%) 51 (48.1%) < 0.001⁎
64 (40.3%) 57 (34.3%)
228 (93.8%) 76 (92.7%) 203 (92.7%) 101 (95.3%)
91 (37.4%) 30 (36.6%) 66 (30.1%) 55 (51.9%)
p < .05.
53
Annals of Diagnostic Pathology 34 (2018) 50–55
U. Agrawal et al.
Fig. 3. Kaplan Meier survival plot shows that patient survival is best when COX2 expression is negative (66.8% cumulative survival at 41 months of follow-up) but with increase in time the survival is almost equal irrespective of COX2 status, falling to 31.0% at the 48th month in COX2 negative individuals. COX2 positive in > 41% tumor cells indicates a bad prognosis with a cumulative survival of 12.1% at 35 months follow-up.
colorectal cancer risk by suppressing the immune response. NSAIDs block endogenous prostaglandin synthesis through inhibition of cyclooxygenase enzymatic activity, both COX1 and COX2 [34]. Similar decrease in bladder cancer risk has been reported with the use of NSAIDs [29]. COX2 specific inhibitors, the coxibs are also now available and act by down-regulating bcl2 (which is a suppressor of apoptosis) [35, 36]. Inhibition of COX2 causes reduction of bladder cancer chemoresistance and increases apoptosis of tumor including cancer stem cells [36].
COX-2 contributes to superficial bladder cancer recurrence [29]. COX2 expression has also been reported to be higher in muscle-invasive bladder cancer in a meta-analysis conducted by Czachorowski et al. However, the same group reported only a weak association with recurrence [22]. The distribution of cases in our centre shows a low incidence of PUNLMP (papillary urothelial neoplasia of unknown malignant potential) cases and of low grade muscle invasive urothelial carcinoma cases. While Tokgoz et al., 2007 [30] report almost 50% low grade tumors in the muscle invasive urothelial carcinomas, Jimenez et al., 2000 [31] have reported very few low grade cases in their cohort. The variation in incidence in different cohorts could also be attributed to the availability of access to advanced health care facility. PUNLMP case may recur and progress. Maxwell et al., 2015 reported a progression of PUNLMP to low grade urothelial carcinoma in 9.5% patients and to high grade urothelial carcinoma in 1.6% patients [32]. In the same year, Zhang et al. reported a recurrence in 19.7% and progression up to pTa in 16.9%. They reported that increased mitosis, tumor multifocality and age at diagnosis were independent prognostic predictors [33]. However, the role of biomarkers in predicting prognosis has not been well studied. However, localization of COX2 in either inflammatory, stromal or malignant cell component and the association of overexpression of COX2 in any one of the component with grade, stage and recurrence has not been explored thoroughly. In the present study the upregulation of the gene and expression of COX2 in inflammatory cells was not found to be associated with stage of the tumor. However, inflammatory cell expression of COX2 was found to be associated with high grade urothelial cancer, possibly reflecting host immune response to high grade tumors. Tumor cell expression of COX2 has been found to be associated with muscle invasion (p = .011) and recurrence (p < .001) in our study. The role of COX2 could not be evaluated in cases of PUNLMP as we had only 2 cases on record. Further evaluation of COX2 in such cases could help to identify its role as a prognostic indicator in early stages. Inhibition of COX2 in aspirin users has been found to decrease
5. Conclusion The association of COX2 with progression of disease, invasion, metastasis and recurrence along with the epidemiological evidence of decreased risk in habitual users of NSAIDs suggests a strong role of COX2 inhibitors in disease management. The present study shows the specific association of tumor cells expressing COX2 with recurrence and invasion of urothelial cancers indicating its role as a marker of aggressive behavior and also a beneficial role for targeting of this enzyme in management of Urothelial cancer patients. Acknowledgements The authors would like to acknowledge the Indian Council of Medical Research (53/4/2008-BMS) for financial support. References [1] Ploeg M, Aben KKH, Kiemeney LA. The present and future burden of urinary bladder cancer in the world. World J Urol 2009;27(3):289–93. [2] Heney NM. Natural history of superficial bladder cancer. Prognostic features and long-term disease course. Urol Clin North Am 1992;19:429–33. [3] Pasin E, Josephson DY, Mitra AP, Cote RJ, Stein JP. Superficial bladder cancer: an update on etiology, molecular development, classification, and natural history. Rev Urol 2008;10(1):31–43. [4] McDougal WS, Shipley WU, Kaufman DS, Dahl DM, Michaelson MD, Zietman AL. Cancer of the bladder, ureter and renal pelvis. In: DeVitaHellmanRosenbergeditors.
54
Annals of Diagnostic Pathology 34 (2018) 50–55
U. Agrawal et al.
[5]
[6]
[7] [8] [9]
[10]
[11] [12]
[13]
[14]
[15]
[16]
[17] [18] [19]
[20]
Neurochem 2000;75:2020–8. [21] Kuwano T, Nakao S, Yamamoto H, Tsuneyoshi M, Yamamoto T, Kuwano M, et al. Cyclooxygenase 2 is a key enzyme for inflammatory cytokine-induced angiogenesis. FASEB J 2004;18:300–10. [22] Czachorowski MJ, AFS Amaral, Montes-Moreno S, Lloreta J, Carrato A, Tardón A, et al. Cyclooxygenase-2 expression in bladder cancer and patient prognosis: results from a large clinical cohort and meta-analysis. PLoS ONE 2012;7(9):e45025. [23] Hill R, Li Y, Tran LM, Dry S, Calvopina JH, Garcia A, et al. Cell intrinsic role of COX2 in pancreatic cancer development. Mol Cancer Ther 2012;11(10):2127–37. [24] Ferrandina G, Lauriola L, Zannoni GF, Fagotti A, Fanfani F, Legge F, et al. Increased cyclooxygenase-2 (COX-2) expression is associated with chemotherapy resistance and outcome in ovarian cancer patients. Ann Oncol 2002;13(8):1205–11. [25] Singh B, Berry JA, Shoher A, Ramakrishnan V, Lucci A. COX-2 overexpression increases motility and invasion of breast cancer cells. Int J Oncol 2005;26(5):1393–9. [26] Cohen BL, Gomez P, Omori Y, Duncan RC, Civantos F, Soloway MS, et al. Cyclooxygenase-2 (COX-2) expression is an independent predictor of prostate cancer recurrence. Int J Cancer 2006;1;119(5):1082–7. [27] Williams CS, Mann M, RN DuBois. The role of cyclooxygenases in inflammation, cancer, and development. Oncogene 1999;18:7908–16. [28] Tomozawa S, Tsuno NH, Sunami E, Hatano K, Kitayama J, Osada T, et al. Cyclooxygenase-2 overexpression correlates with tumour recurrence, especially haematogenous metastasis, of colorectal cancer. Br J Cancer 2000;83(3):324–8. [29] Okajima E, Uemura H, Ohnishi S, Tanaka M, Ohta M, Tani M, et al. Expression of cyclooxygenase-2 in primary superficial bladder cancer tissue may predict risk of its recurrence after complete transurethral resection. Aktuelle Urol 2003;34(4):256–8. [30] Tokgoz Husnu, Turkolmez Kadir, Resorlu Berkan, Kose Kenan, Tulunay Ozden, Beduk Yasar. Pathological staging of muscle invasive bladder cancer: is substaging of pT2 tumors really necessary? Int Braz J Urol 2007;33(6):777–84. [31] Jimenez RE, Gheiler E, Oskanian P, Tiguert R, Sakr W, Wood Jr. DP, et al. Grading the invasive component of urothelial carcinoma of the bladder and its relationship with progression free survival. Am J Surg Pathol 2000;24(7):980–7. [32] Maxwell JP, Wang C, Wiebe N, Yilmaz A, Trpkov K. Long-term outcome of primary papillary urothelial neoplasm of low malignant potential (PUNLMP) including PUNLMP with inverted growth. Diagn Pathol 2015;10(3). [33] Zhang XK, Wang YY, Chen JW, Qin T. Bladder papillary urothelial neoplasm of low malignant potential in Chinese: a clinical and pathological analysis. Int J Clin Exp Pathol 2015;8(5):5549–55. [34] Xu XC. COX-2 inhibitors in cancer treatment and prevention, a recent development. Anticancer Drugs 2002;13(2):127–37. [35] Steinbach G, Lynch PM, Phillips RK, Wallace MH, Hawk E, Gordon GB, et al. The effect of celecoxib, a cyclooxygenase-2 inhibitor, in familial adenomatous polyposis. N. Engl. J. Med. 2000;342:1946–52. [36] Kurtova AV, Xiao J, Mo Q, Pazhanisamy S, Krasnow R, Lerner SP, et al. Blocking PGE2-induced tumour repopulation abrogates bladder cancer chemoresistance. Nature 2015;517:209–13.
Cancer, Principles and Practice of Oncology. 9th editionPhiladelphia: Lippincott Williams & Wilkins; 2011. p. 1192–211. Zlotta AR, van Vooren JP, Huygen K, et al. What is the optimal regimen for BCG intravesical therapy? Are six weekly instillations necessary? Eur Urol 2000;37(4):470–7. Lebret T, Bohin D, Kassardjian Z, Herve JM, Molinie V, Barre P, et al. Recurrence, progression and success in stage Ta grade 3 bladder tumors treated with low dose bacillus Calmette-Guerin instillations. J Urol 2000;163(1):63–7. Grivennikov SI, Greten FR, Karin M. Immunity, inflammation, and cancer. Cell 2010;140(6):883–99. Gajewski TF, Schreiber H, Fu Y. Innate and adaptive immune cells in the tumor microenvironment. Nat Immunol 2013;14:1014–22. Greenhough A, Smartt HJM, Moore AE, Roberts HR, Williams AC, Paraskeva C, et al. The COX-2/PGE2 pathway: key roles in the hallmarks of cancer and adaptation to the tumour microenvironment. Carcinogenesis 2009;30(3):377–86. Vogel LK, Sæbø M, Høyer H, Kopp TI, Vogel U, Godiksen S, et al. Intestinal PTGS2 mRNA levels, PTGS2 gene polymorphisms, and colorectal carcinogenesis. PLoS ONE 2014;9(8):e105254. Allaj V, Guo C, Nie D. Non-steroid anti-inflammatory drugs, prostaglandins, and cancer. Cell Biosci 2013;3:8. Asting AG, Carén H, Andersson M, Lönnroth C, Lagerstedt K, Lundholm K. COX-2 gene expression in colon cancer tissue related to regulating factors and promoter methylation status. BMC Cancer 2011;11(238). Roelofs HMJ, te Morsche RHM, van Heumen BWH, Nagengast FM, Peters WHM. Over-expression of COX-2 mRNA in colorectal cancer. BMC Gastroenterol 2014;14(1). Ferrandina G, Lauriola L, Zannoni GF, et al. Expression of cyclooxygenase-2 (COX2) in tumour and stroma compartments in cervical cancer: clinical implications. Br J Cancer 2002;87(10):1145–52. Ferrandina G, Ranelletti FO, Legge F, Gessi M, Salutari V, Distefano MG, et al. Prognostic role of the ratio between Cyclooxygenase-2 in tumor and stroma compartments in cervical cancer. Clin Cancer Res 2004;10:3117–23. Kim M-J, Kim H-S, Lee S-H, Yang Y, Lee M-S, Lim J-S. NDRG2 controls COX-2/ PGE2-mediated breast cancer cell migration and invasion. Mol Cells 2014;37(10):759–65. Jana D, Sarkar DK, Ganguly S, Saha S, Sa G, Manna AK, et al. Role of cyclooxygenase 2 (COX-2) in prognosis of breast cancer. Ind J Surg Oncol 2014;5(1):59–65. Michalowski J. COX-2 inhibitors: cancer trials test new uses for pain drug. JNCI J Natl Cancer Inst 2002;94(4):248–9. Sauter G, Algaba F, Amin MB, et al. Non-invasive urothelial tumours. In: Eble JN, Sauter G, Epstein JI, Sesterhenn IA, editors. World Health Organization Classification of Tumours: Pathology and Genetics of Tumours of the Urinary System and Male Genital Organ. Lyon, France: IARC Press; 2004. p. 110–23. Fiebich BL, Mueksch B, Boehringer M, Hull M. Interleukin-1beta induces cyclooxygenase-2 and prostaglandin E(2) synthesis in human neuroblastoma cells: involvement of p38 mitogen-activated protein kinase and nuclear factorkappaB. J
55