A Prospective Study of the Role of Inflammation in Bladder Cancer

Abstract

Introduction

To examine the role of inflammation in bladder cancer, we assessed the relationship between a systemic inflammation prognostic score (modified Glasgow Prognostic Score, mGPS), the tumor inflammatory cell infiltrate as measured by the Klintrup-Makinen score and tumor necrosis with cancer specific survival in patients with bladder cancer.

Materials and Methods

The cohort consisted of 68 bladder cancer patients, 47 with localised disease and 21 with muscle invasive disease. The mGPS response was constructed by measuring C-reactive protein and albumin concentrations and the Klintrup-Makinen score was evaluated histologically for the local inflammatory response. Pathological parameters such as grade, T stage and tumor necrosis were also assessed.

Results

Median follow was 47 months and 24 patients died of their disease. On univariate analysis, T stage (p < 0.001), grade (p < 0.001) and mGPS (p = 0.002) were significant predictors of cancer specific survival. On multivariate analysis, T stage (hazard ratio 5.98, 95% confidence interval 3.18–11.24, p < 0.001) and mGPS (hazard ratio 1.78, 95% confidence interval 1.09–2.9, p = 0.02) were significant independent predictors of cancer specific survival.

Conclusion

A preoperative systemic inflammatory response is an independent predictor of poor cancer specific survival in patients with bladder cancer.

Introduction

Worldwide there are approximately 360,000 new cases of bladder cancer diagnosed each year. In the UK alone, bladder cancer is the fifth most common solid non-cutaneous malignancy with approximately 10,000 new cases and 5,000 deaths annually. Overall survival is variable as only half of patients diagnosed survive 5 years [1].

There has been a long standing interest in identifying those patients most at risk of disease progression and ultimately dying from their disease. Ideally, a factor or combination of factors that could clearly stratify patients into those who do not progress and those that progress and are at higher risk of dying from their cancer would be highly beneficial. Currently, the TNM stage and tumor grade are the most widely used tools to predict survival but could be improved upon.

It is now established that disease progression in cancer patients is not solely determined by the tumor characteristics but also by the host response. There is increasing evidence that both local and systemic inflammatory responses play an important role in the progression of various solid tumors [2, 3].

Recent evidence suggests that intensity of local inflammatory infiltrate within the tumor bed predicts prognosis: a pronounced lymphocytic infiltration in colorectal cancer is associated with improved survival [4, 5, 6]. Also, quantifying the degree of infiltration by lymphocyte subsets such as CD8+ and CD4+ T cells provides prognostic information in a various tumor types [6, 7, 8] including bladder cancer [9]. The process of assessing lymphocyte infiltration is time consuming and has not been adopted into routine clinical practice. It is therefore of interest that Klintrup et al. [10] have reported a simplified method of assessing the inflammatory cell infiltrate at the tumor margin shown on routine hematoxylin and eosin stained sections, that tumor inflammatory infiltrate, including all white cell types, can be graded high or low grade, with a high grade infiltrate being associated with improved survival in colorectal cancer.

In addition, increasing evidence supports a role of the systemic inflammatory response, indicated by elevated levels of C-reactive protein (CRP) being an independent predictor of survival in patients with a variety of common solid tumors including gastrointestinal, lung, renal, prostate and bladder cancers [11, 12, 13, 14, 15, 16]. A retrospective analysis of patients undergoing radical cystectomy has demonstrated that CRP in combination with other pathological parameters offers additional prognostic information in these patients but some of these parameters are collated post procedure [17]. The modified Glasgow Prognostic Score (mGPS), incorporates CRP and albumin serum levels preoperatively [18]. mGPS score provides additional prognostic information in patients with various solid malignancies including lung, gastroesophageal, renal and colorectal cancers [19, 20, 21, 22].

The prognostic role of tumor necrosis is well established in malignancies such as renal, breast, lung and colorectal cancers [23, 24, 25, 26]. However, there are few reports on the prognostic role of necrosis in bladder cancer.

The aim of this study was to assess relationships between a systemic inflammation prognostic score (mGPS), the local tumor inflammatory cell infiltrates, tumor necrosis and cancer specific survival in patients with bladder cancer.

Materials and Methods

Patients with bladder cancer diagnosed in the North Glasgow NHS Trust were included. Patients were staged pathologically according to the TNM classification and graded according to the World Health Organisation/International Society of Urological Pathology criteria [27]. The Research Ethics Committee of North Glasgow NHS Trust approved the study.

Diagnostic hematoxylin and eosin sections from pathology archives were reviewed. A minimum of 3 slides from the deepest area of tumor invasion were reviewed and were scored according to the Klintrup-Makinen criteria (K/M) [10]. This method is based on scoring inflammation at the deepest point of invasion identified from the 3 slides. A 4-point scale was used. A score of 0 was given for no increase of the inflammatory cells at the invasive margin; a score of 1 denoted a mild and patchy increase of inflammatory cells. Score 2 was assigned when inflammatory cells formed a band-like infiltrate at the invasive margin. A score of 3 was given when a prominent inflammatory reaction formed a cup-like zone at the margin. Scores of 0 and 1 were combined (low grade inflammation) and scores of 2 and 3 combined (high grade inflammation). Scoring was supervised by a pathologist (J.J.G.).

Preoperative systemic inflammatory response was assessed using the mGPS [18]. Patients with both elevated CRP (> 10mg/l) and hypoalbuminemia (< 35g/l) scored 2. Patients in whom both were normal scored 0. Patients with elevated CRP alone were scored as 1 while those with hypoalbuminemia alone were scored as 0.

The presence or absence of tumor necrosis was evaluated on histological sections under the supervision of a pathologist (J.J.G.).

Statistical analysis was undertaken using SPSS. Disease specific survival rates were generated using the Kaplan-Meir method. The log rank test was utilised to compare significant differences between subset groups using univariate analysis. Multivariate analysis was carried out based on the results of the univariate analysis. Multivariate Cox regression analysis was performed to identify factors independently associated with disease specific death. A stepwise backward procedure was utilised to ascertain which of the variables had a significant independent relationship with survival.

Results

Sixty-eight patients were studied, 47 with localised disease and 21 with muscle invasive disease. Median age at diagnosis was 72 years (range 43–93 years). Median follow-up was 47 months (range 1.2–201 months). Forty patients had recurrence and of these, 28 patients had evidence of progression and 24 patients died of their disease.

χ² demonstrated that the mGPS was positively correlated with tumor stage, grade and progression (p = 0.033, 0.007 and 0.033, respectively) (table 1). There was also a positive correlation between necrosis and local inflammatory infiltrate (p = 0.027) (table 1). In a subset analysis looking at localised and muscle invasive cancer disease separately, the only correlation was a negative one with mGPS and tumor necrosis in those with localised disease (p = 0.025). There was no significant difference in mGPS, tumor necrosis and the local inflammatory response when comparing those with muscle invasive and localised disease.

Table 1.

Interrelationships between clinicopathological characteristics of patients with bladder cancer

	Numbers	Sex	Grade	T stage	Tumor necrosis	Local inflammatory cell infiltrate	mGPS	Recurrence	Progression
Age (>65/<65)	25/43	0.037	0.223	0.037	0.962	0.82	0.775	0.17	0.095
Sex (male/female)	46/22		0.546	0.593	0.539	0.147	0.653	0.975	0.975
Grade (1/2/3)	15/28/25			<0.001	0.445	0.207	0.007	0.021	<0.001
T stage (Ta/T1/T2/T3/T4)	34/13/20/1/0				0.376	0.311	0.033	<0.001	<0.001
Tumour Necrosis (absence/presence)	22/46					0.027	0.212	0.31	0.11
Local inflammatory cell infiltrate (low/high)	59/9						0.444	0.61	0.098
mGPS (0/1/2)	44/13/11							0.436	0.033
Recurrence (no/yes)	28/40								<0.001
Progression (no/yes)	40/28

On univariate analysis, T stage (p < 0.001, fig. 1), grade (p < 0.001, fig. 2) and mGPS (p = 0.002, fig. 3) were significant predictors of cancer specific survival whilst necrosis and the local inflammatory response did not show significance (p = 0.286 and 0.253, respectively) (table 2). On multivariate analysis of the significant individual covariates T stage (hazard ratio 5.98, 95% confidence interval 3.18–11.24, p < 0.001) and mGPS (hazard ratio 1.78, 95% confidence interval 1.09–2.90, p = 0.02) were significant independent predictors of cancer specific survival.

Fig. 1.

A Kaplan-Meier plotted for the T stage against disease specific survival (log rank, p < 0.001).

Fig. 2.

A Kaplan-Meier plotted for the nuclear grade against disease specific survival (log rank, p < 0.001).

Fig. 3.

A Kaplan-Meier plotted for the mGPS score against disease specific survival (log rank p = 0.002).

Table 2.

The relationship between clinicopathological characteristics, tumou necrosis, local inflammatory response, systemic inflammatory response and cancer specific survival in patients undergoing treatment for bladder cancer

	Numbers	Univariate analysis, p	Multivariate analysis, p	Multivariate analysis, HR
Age (<65/>65)	25/43	0.024
Sex (male/female)	46/22	0.756
Grade (1/2/3)	15/28/25	<0.001
T stage (Ta/T1/T2/T3/T4)	34/13/20/1/0	<0.001	<0.001	5.98 (3.18–11.24)
Tumor Necrosis (absence/presence)	22/46	0.286
Local inflammatory cell infiltrate (low/high)	59/9	0.253
mGPS (0/1/2)	44/13/11	0.002	0.02	1.78 (1.09–2.90)

Discussion

Results from the present study demonstrate, for the first time to our knowledge that an elevated mGPS independently correlates to a poor cancer specific survival in those with bladder cancer. We have also demonstrated that an elevated mGPS is directly associated with tumor stage, grade and progression.

Previous studies have demonstrated that tumor necrosis [23, 24, 25, 26] and the local inflammatory response [10, 11] play a prognostic role in various malignancies. It has also been reported that quantifying the degree of infiltration by lymphocyte subsets provides prognostic information in bladder cancer [9]. It was therefore of interest that a simplified assessment of the local inflammatory cell infiltrate [10] which is not as time consuming as quantifying lymphocyte subsets has been shown to prognostic in colorectal cancer [10, 11]. In this study, there was a positive correlation with necrosis and the local inflammatory infiltrate but neither of these pathological parameters showed any significance to cancer specific survival, this could reflect the number of patients in this cohort.

The mGPS utilises objective, well established serum based variables to predict cancer specific survival in patients undergoing treatment for bladder cancer. The variables used are common ones and offer the benefit of being objective and obtainable. These results support the evaluation and introduction of the inflammation based (mGPS) scoring system as an independent predictor of poor cancer specific survival in bladder cancer. Limitations of this study include patient numbers. Given the different prognosis between non-muscle invasive disease and muscle invasive disease, ideally these should be analysed separately, but this is not possible given the low numbers in each group.

Further work is required to see if the mGPS can be used as a guide for adjuvant treatment in bladder cancer.