SEARCH

SEARCH BY CITATION

Keywords:

  • C-reactive protein;
  • bladder cancer;
  • radiotherapy;
  • inflammatory reaction;
  • biological marker;
  • prognosis

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

OBJECTIVE

To investigate the effect of C-reactive protein (CRP) level on the prognosis of patients with muscle-invasive bladder cancer treated with chemoradiotherapy (ChRT), as it is increasingly recognized that the presence of a systemic inflammatory response is associated with poor survival in various malignancies.

PATIENTS AND METHODS

The clinical records of 88 patients with bladder urothelial carcinoma (cT2–4 N0M0) treated with ChRT were reviewed retrospectively. ChRT comprised external beam radiotherapy to the bladder (40 Gy) with two cycles of cisplatin (50–100 mg) at 3-week intervals. Elevated CRP was defined as >0.5 mg/dL. The survival rate was calculated using the Kaplan-Meier method, and a multivariate analysis was used to identify significant factors associated with prognosis, using a Cox proportional hazards model.

RESULTS

During the median (range) follow-up of 33 (3–117) months, 19 patients died from bladder cancer; the 5-year cancer-specific survival (CSS) rate was 73%. Ten patients had a high CRP level before ChRT (≥0.5 mg/dL) and their CSS rate was significantly worse than that in the remaining patients (= 0.003). Multivariate analysis showed that CRP and cT stage were independent prognostic indicators for CSS, with a hazard ratio of 1.80 (95% confidence interval 1.01–2.97; P = 0.046). Among 10 patients in those with elevated CRP the CRP levels became normal after ChRT in six, of whom all but one was alive with no evidence of recurrence or metastasis during the follow-up. By contrast, all four with no CRP normalization after ChRT died within 2 years.

CONCLUSIONS

To our knowledge this is the first study to report that elevation of CRP before treatment predicts a poor prognosis in patients with muscle-invasive bladder cancer who are receiving ChRT. Furthermore, failure of CRP levels to normalize after ChRT was associated with extremely poor survival.


Abbreviations
ChRT

chemoradiotherapy

CRP

C-reactive protein

CSS

cancer-specific survival

CR

complete response

UC

urothelial carcinoma

TURBT

transurethral resection of the bladder tumour.

INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

Bladder cancer is the second most common genitourinary cancer in Japan; 12 000 new patients are diagnosed and 5000 patients die each year [1]. A third of patients with bladder cancer are diagnosed as having muscle-invasive tumour and about half of them treated with the standard therapy of radical cystectomy eventually develop disease progression [2]. We have treated patients with muscle-invasive bladder cancer by chemoradiotherapy (ChRT) to control this locally invasive disease, and reported favourable outcomes with this treatment [3]. However, in some patients the disease progressed after ChRT within a few years. Therefore, identifying a prognostic factor for this would allow a better therapeutic approach to patients with muscle-invasive bladder cancer.

In terms of muscle-invasive bladder cancer, some tissue markers, e.g. various oncogenes, tumour-suppressor genes, microvessel density and angiogenic inhibitors, were reported to be able to predict the response to chemotherapy [4–6]. However, there are no reports of a prognostic biomarker for predicting the response patients with muscle-invasive bladder cancer receiving ChRT.

It is increasingly recognized that the presence of a systemic inflammatory response is associated with poor survival in various malignancies [7–9]. In genitourinary cancers, we reported that a high C-reactive protein (CRP) level is a poor prognostic indicator in patients with RCC and upper urinary tract urothelial carcinoma (UC) [10,11]. In bladder cancer, an elevated CRP level was associated with poor survival in a group of patients with heterogeneous staging and treatment [12]. The aim of the present study was to examine the effect of serum CRP levels on the prognosis of patients with muscle-invasive bladder cancer receiving ChRT.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

From 1997 to 2006, in all, 108 consecutive patients with muscle-invasive bladder UC with no lymph node involvement and distant metastasis (cT2–4 N0M0) were treated with ChRT in our institution. The clinical stage according to the 1997 International Union Against Cancer classification [13] was determined based on the findings of transurethral resection of the bladder tumour (TURBT) before ChRT, and CT and bone scans; histological grade was determined according to the WHO grading system [14]. The CRP level before ChRT was defined as that estimated at dawn before TURBT; this value was unavailable in 19 patients and untreated lung cancer was present in one patient. Consequently, 88 patients comprised the study group; at diagnosis, none of these patients had any inflammatory disease.

ChRT comprised external beam RT (40 Gy) and cisplatin chemotherapy, as described previously [3]. In principle, radiation to the bladder was given to a dose of 40 Gy with the daily fractional dose of 1.8 or 2 Gy in a 5-day week. Cisplatin was administered systemically (20 mg/body for 5 days) or intra-arterially (100 mg/body) during weeks 1 and 4 of RT. The radiation dose was reduced in three patients (3%) due to acute bladder toxicity (median total dose 40 Gy; range 15.0–59.3). The dose of cisplatin was reduced depending on the patient’s renal function (median total dose 155 mg, range 15–200).

From 1997 to 2000 we used ChRT as neoadjuvant therapy followed by radical cystectomy. In emergency cases with advanced age or where radical cystectomy was a high risk, partial cystectomy was used. Based on the outcome of these patients with a preserved bladder we refined the ChRT protocol as an induction therapy. Thus, since 2000, partial cystectomy with pelvic lymph node dissection of curative intent was used according to the tumour status and patient’s general status after ChRT. Consequently, the treatment after ChRT was radical cystectomy, partial cystectomy and TURBT in 54, 12 and 14 patients, respectively. The pathological response to ChRT from specimens obtained in subsequent surgical procedures was as follows; a complete response (CR) was defined as no viable tumour cells, and no response as the presence of viable cells; of the 88 patients, 37 (42%) had a pathological CR.

Serum CRP was measured by latex agglutination using CRP-L kit (Mitsubishi Kagaku Iatron Co. Ltd, Tokyo, Japan). We assigned patients with a CRP level before ChRT of >0.5 mg/dL to the elevated CRP group. Furthermore, in this group we defined patients with the ‘normal’ level of CRP (<0.5 mg/dL) after the ChRT within 1 month as CRP responders.

Associations between clinicopathological features and CRP level were analysed using the chi-square test. Cancer-specific survival (CSS) was calculated and analysed using the Kaplan-Meier method and log-rank test. Multivariate analysis used the Cox proportional hazards model, with factors before ChRT including age, sex, clinical (cT) stage, tumour grade, tumour structure, application of debulking TURBT, and CRP level. For the statistical analysis patients were divided in to those aged > or <70 years (the median age of the group). The cT stage, tumour grade and tumour structure were divided into two groups, i.e cT2 vs cT3/4, G2 vs G3, and papillary vs non-papillary, respectively. Subsequent surgery was divided into two groups, radical cystectomy vs bladder-conservative therapy. The duration of CSS was defined as the interval from the date of diagnosis of muscle-invasive bladder cancer. In all statistical tests the significance level was determined at 0.05.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

The patients’ characteristics are described in Table 1. During the median (range) follow-up of 38 (3–117) months, 19 patients (22%) died from bladder cancer. The 5-year CSS rate was 74%. Radical cystectomy or bladder-conservative therapy as subsequent surgery was not associated with CSS (= 0.88). Ten patients (11%) had an elevated CRP level; their cT stage was significantly higher (= 0.033). Five of these 10 patients, and 14 of the 78 (18%) in the ‘normal’ CRP group, died from disease (= 0.020). In the former group, five died from their disease, one from other cause with no recurrence of bladder cancer, and four were alive with no recurrence. As shown in Fig. 1, the CSS of the elevated CRP group was significantly worse than that of the ‘normal’ CRP group (= 0.003), the 5-year CSS rates being 50% and 77%, respectively. Of the other factors, cT was associated with poorer CSS rate (= 0.002; Table 2). Multivariate analysis of the factors before ChRT showed that CRP and cT stage were independent prognostic indicators for CSS, with a hazard ratio (95% CI) of 1.80 (1.01–2.97) (P = 0.046).

Table 1.  The relationship between CRP levels and clinicopathological characteristics of patients treated with ChRT
VariableCRPP
AllNormalElevated
  1. Data are expressed as the median (25th−75th percentiles) or number. PC, partial cystectomy; RC, radical cystectomy.

No. of patients887810 
Age, years70 (63–75)70 (64.5–75)63.5 (60–74)0.190
Male63567 
Female252230.800
cT stage
 256533 
 3/43221/46/10.033
Grade
 210 73 
 3787170.680
Debulking TURBT before ChRT    
 Yes44422 
 No443680.089
Tumour structure
 Papillary25232 
 Not papillary635580.720
Treatment after ChRT
 None/TURBT/PC 8/14/12 6/13/122/1/0 
 RC544770.540
Pathological response
 CR37334 
 No CR433940.880
image

Figure 1. Kaplan-Meier analysis of CSS stratified by CRP levels before ChRT in patients with muscle-invasive bladder carcinoma. The number of patients at risk is shown on the horizontal axis at 0 and 60 months.

Download figure to PowerPoint

Table 2.  The characteristics of patients with muscle-invasive bladder cancer treated with ChRT; univariate and multivariate analysis of CSS
Variable, categoryN patientsUnivariate PMultivariate
HR (95% CI)P
  1. HR, hazard ratio.

Age, years, ≥70 vs <7045/430.70 
Male vs female63/250.47 
cT stage, cT3/4 vs cT232/560.00211.87 (1.17–3.10)0.010
Grade, G3 vs G278/100.91 
Debulking TURBT, no vs yes44/440.42 
Tumour structure, non-papillary vs papillary63/250.40 
CRP level, ≤0.5 vs <0.5 mg/dL10/780.00301.80 (1.01–2.97)0.046

The detailed clinicopathological characteristics of the patients with an elevated CRP are shown in Table 3. In six of the 10 patients the CRP levels became normal after ChRT. In these six CRP responders, all but one were alive with no recurrence or metastasis during the median (range) follow-up of 30 (9–111) months. By contrast, all four who did not respond (no normal CRP level after ChRT) died from the disease within 2 years, due to distant or lymph node metastasis. The prognosis of those not responding was worse than that of the CRP responders (= 0.047).

Table 3.  Clinicopathological characteristics, pathological response to ChRT and prognosis of the 10 patients with a high CRP level before ChRT
Patient/ age/sexTumour structureGradecT stageCRP level* before/afterPathological responseSubsequent treatmentRFS, monthsSite of disease progressionOutcomeCSS, months
  • *

    mg/dL, before and after ChRT; RFS, recurrence-free survival, excluding superficial bladder tumour recurrence; P, papillary; N, not papillary; RC, radical cystectomy; DOD, dead from disease; DOO, death from other cause; NED, no evidence of disease.

1/63/FN320.5/0.3CRRC111NED111
2/77/MN332.0/0.2CRRC27DOO27
3/73/MP221.0/0.1CRTURBT49NED49
4/62/MP230.5/0.1No CRRC34NED34
5/64/MN330.5/0.2No CRRC26NED26
6/52/MN330.8/0.1CRRC3LocalDOD 9
7/81/MN343.4/1.5No CRRC10BoneDOD16
8/68/MN222.7/5.2No CRNone5Iliac lymph nodeDOD5
9/60/FN332.6/3.7No CRRC5Iliac lymph nodeDOD6
10/60/FN331.1/2.8No CRNone3PeritoneumDOD3

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

To our knowledge this is the first study to show the prognostic effect of CRP level in patients with muscle-invasive bladder cancer treated with ChRT. In the present study, CRP and cT stage before ChRT predicted a poor prognosis in these patients. This is consistent with our previous report showing the prognostic effect of CRP level in patients with upper urinary tract UC [10].

The presence of a systemic inflammatory response is reported to be associated with a poor prognosis in patients with various malignancies. Recently, UC was shown to synthesise interleukin-6, functioning as an autocrine growth factor and enhancing the proliferation of carcinoma cells [15–17]. The correlation between elevated CRP levels and a poor prognosis might reflect the prognostic value of tumour produced interleukin-6, an inducer of CRP production in the liver.

In the present study it was interesting that, apart from being a prognostic indicator, CRP could be a useful marker for the clinical course of patients with muscle-invasive bladder UC treated with ChRT. In the high-CRP group CRP responders had a more favourable outcome than those whose CRP level did not respond. Thus return to a normal CRP level might be used as a surrogate endpoint of clinical outcome. The failure of CRP levels to become normal after treatment indicated further disease progression and the need for adjuvant therapy. Therefore, CRP could be a practical adjunct in stratification before and after treatment of patients with muscle-invasive bladder cancer.

The proportion of patients with muscle-invasive bladder cancer and an elevated CRP level was low, at 11%. Moreover, in assessing CRP elevation, the presence of other inflammatory disease should be considered because CRP is a nonspecific inflammatory marker. However, measuring the CRP level is simple and of low cost. Therefore, despite this weak sensitivity, CRP could be routinely measured as a practical clinical marker in patients with muscle-invasive bladder cancer treated with ChRT.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES