Tumor Biology

, Volume 34, Issue 1, pp 39–45

Increased platelet count is an indicator of metastasis in patients with nasopharyngeal carcinoma

Authors

  • Jin Gao
    • Department of Radiation OncologyAnhui Provincial Hospital
  • Hong-Yan Zhang
    • Department of Radiation OncologyAnhui Provincial Hospital
    • Department of Radiation Oncology, Cancer CenterSun Yat-Sen University
Research Article

DOI: 10.1007/s13277-012-0508-y

Cite this article as:
Gao, J., Zhang, H. & Xia, Y. Tumor Biol. (2013) 34: 39. doi:10.1007/s13277-012-0508-y

Abstract

This study aims to evaluate the impact of pretreatment thrombocytosis on survival in patients with nasopharyngeal carcinoma (NPC). The data of 1,582 patients with NPC, who underwent definitive treatment between 2003 and 2004, were retrospectively reviewed. The correlation between the clinicopathological variables and the platelet count was analyzed. The prognostic significance of thrombocytosis, together with various clinicopathological factors, was evaluated by univariate and multivariate analyses. Platelet count showed significant correlation with gender, clinical stage, and T stage in univariate analysis. There was poorer 5-year disease-specific survival (DSS) in the patients with thrombocytosis than in those without thrombocytosis (70 vs. 78 %; p = 0.001) and poorer metastases-free survival (MFS) (81 vs. 88 %; p = 0.006). Univariate and multivariate analyses showed that thrombocytosis was an independent prognostic factor for MFS and DSS. Thrombocytosis is a useful predictor of metastasis and poor prognosis in patients with NPC.

Keywords

ThrombocytosisPlatelet countNasopharyngeal carcinomaSurvivalMetastasis

Introduction

Nasopharyngeal carcinoma (NPC) is a unique malignancy showing a distinct racial and geographical distribution. The highest incidence rates are found in southern Chinese. Globally, there were 84,400 cases of NPC and 51,600 deaths in 2008 [1]. Radiotherapy is the most important treatment modality. Although NPC is markedly radiosensitive, there is a significant rate of local failures and distant metastases [2]. Patients with recurrent disease or distant metastasis after treatment have a grim prognosis. To accurately estimate the prognosis is crucial for optimizing treatment strategies for cancer patients. The present TNM staging system is the most popular and used worldwide. However, patients with the same clinical stage often present different clinical outcomes, indicating that the clinical staging is insufficient for precisely predicting a prognosis of NPC. There is increasing evidence that several molecular biomarkers, such as Met [3] and high centromere protein H expression [4], could be predictive of poor survival in patients with NPC. However, these biomarkers are not used routinely because they are often time-consuming to measure. An easily inexpensive marker that could better predict the outcome of patients with NPC would be of substantial value for clinicians involved in the treatment of NPC.

Thrombocytosis is commonly observed in neoplastic diseases. Elevated levels of platelet counts may be associated with tumor progress. Various studies have showed that thrombocytosis is associated with poor prognosis in ovarian cancer [5], renal cell carcinoma [6, 7], colorectal cancer [8], gastric cancer [9], and endometrial carcinomas [10]. There were only a few papers on the prognostic value of thrombocytosis in squamous cell carcinoma of the head and neck. Lu et al. reported that pretreatment thrombocytosis was an independent predictor of shorter survival in patients with oral squamous cell carcinoma [11]. However, there is no report about the possible role of platelet count in patients with NPC. The objective of our study was to evaluate whether the platelet levels could be useful in evaluating a prognosis in NPC patients.

Materials and methods

Population

A total of 1,679 hospitalized patients underwent radiotherapy (RT) for NPC at the Cancer Center of Sun Yat-Sen University between 2003 and 2004. Eligibility criteria were as follows: (1) histological confirmation of NPC, (2) firstly receiving radical RT, and (3) no obvious signs of inflammatory disease. Exclusion criteria included presence of distant metastasis, autoimmune disease, and history of blood transfusion. Thus, a total of 1,582 patients were included in this study. There were 1,187 male patients and 395 females, and the ratio of men to women was 3:1. The median patient age was 45 years (range, 11–78 years). The median follow-up for the whole group was 66 months (range, 2–93 months). This study was approved by the Institutional Review Board of Sun Yat-Sen University Cancer Center.

Pretreatment evaluation included a complete medical history and physical examination, complete blood count and biochemical profile, endoscopic examination of the nasopharynx, computed tomography and/or magnetic resonance imaging (MRI) of the head-and-neck region, type B ultrasound of the abdominal, chest X-ray, and emission computed tomography bone scan. Computed tomography and/or MRI were essential for disease staging before treatment, and the disease of all patients was staged prospectively according to the 2002 American Joint Committee on Cancer staging classifications [12].

Platelet counts were measured at baseline for all patients. The prognostic impact of platelet counts measured at baseline was evaluated by evaluating the correlation between survival time and platelet counts. Platelet count of >300 × 109/L was defined as thrombocytosis, according to the normal range of platelet count of our institution. In our institution, the normal platelet count ranges from 100 to 300 × 109/L.

Treatment

All patients had been treated with definitive-intent RT or combined with neoadjuvant concomitant chemotherapy. Technique ranged from conventional two-dimensional conformal radiotherapy (2D-CRT) to three-dimensional conformal radiotherapy (3D-CRT) or intensity-modulated radiotherapy (IMRT). 2D-CRT consisted of opposing lateral facial–cervical fields to cover the nasopharynx and upper cervical lymphatic drainage region, with one lower anterior cervical field to cover the lower cervical region. After 36 to 40 Gy, opposing lateral preauricular fields were used for the primary region, and anterior split neck fields were used for the cervical region. The primary tumor was irradiated to a dose of 66 to 78 Gy. For 3D-CRT, the total prescribed dose was 66–72 Gy to the gross tumor volume of nasopharynx (GTVnx), 60 to 70 Gy to the region involved by the metastatic lymph nodes (GTVnd), 60 Gy to CTV-1 (the GTVnx and an additional 5- to 10-mm margin ), and 50–54 Gy to the prophylactic irradiating region (CTV-2). For IMRT, the target definition and delineation were the same as described above for 3D-CRT. The prescription dose was 68 Gy to the GTVnx of nasopharynx, 60 to 64 Gy to the GTVnd of neck, 60 Gy to the CTV-1, and 54 Gy to CTV-2. For chemotherapy, of the 1,064 patients with stage III/IV disease, 829 (77.9 %) patients received neoadjuvant concomitant chemotherapy using platinum-based regiments.

Statistical analysis

All analyses were performed using SPSS software, version 13.0 (SPSS, Chicago, IL, USA). Platelet counts were compared using the Mann–Whitney U test. The association between pretreatment thrombocytosis and each clinical factor was examined with a chi-square test. Disease-specific survival (DSS) was calculated as the time from the start of RT to death; only deaths due to disease progression or treatment-related complications were counted. Local regional recurrence-free survival (LRFS) and metastases-free survival (MFS) were calculated as the time from the start of RT to locoregional or distant failure, respectively. The survival was assessed according to the Kaplan–Meier method. Log-rank test was used to compare the survival curves. Multivariate analyses with the Cox proportional hazards model were used to test the independent significance of different variables. Differences were considered statistically significant for values of p < 0.05.

Results

Patient’s clinical characteristics

Patient and disease characteristics are shown in Table 1. Histological examination showed that 99.7 % of the patients had World Health Organization type II or III disease, and 0.3 % of patients had adenocarcinoma. Seventy-five percent of patients were males, and median age was 45 years. Platelet (PLT) count before RT ranged from 60 to 546 × 109/L, with a mean value of 227 × 109/L. Of 1,582 patients, 187 (11.8 %) demonstrated a PLT count of >300 × 109/L before RT.
Table 1

Clinical characteristics data for patients

Characteristic

Case (%)

Age (years)

 ≤45

817 (51.6)

 >45

765 (48.4)

Gender

 Male

1,187 (75.0)

 Female

395 (25.0)

Clinical stage

 I

77 (4.9)

 II

441 (27.9)

 III

693 (43.8)

 IV

371 (23.5)

T stage

 T1

263 (16.6)

 T2

507 (32.0)

 T3

505 (31.9)

 T4

307 (19.4)

N stage

 N0

384 (24.3)

 N1

619 (39.1)

 N2

504 (31.9)

 N3

75 (4.7)

Chemotherapy

 No chemotherapy

753 (47.6)

 Neoadjuvant

239 (15.1)

 Concurrent

590 (37.3)

Relationship between platelet count and clinical characteristics

The relationship between thrombocytosis and clinicopathologic factors is shown in Table 2. Thrombocytosis was significantly related with sex (p = 0.006), T stage (p < 0.001), and clinical stage (p = 0.001). There was no association of thrombocytosis with patient age and N stage.
Table 2

Relationships between platelet count and clinicopathologic factors

Variables

≤300 × 109/L

>300 × 109/L

P value

N = 1,395 (%)

N = 187 (%)

Sex

 Male

1,062 (76.1)

125 (66.8)

0.006

 Female

333 (23.9)

62 (33.2)

Age (years)

 ≤45

710 (50.9)

107 (57.2)

0.10

 >45

685 (49.1)

80 (42.8)

Clinical stage

 I

70 (5.0)

7 (3.7)

0.001

 II

405 (29.1)

36 (19.3)

 III

613 (43.9)

80 (42.8)

 IV

307 (22.0)

64 (34.2)

T stage

 T1

241 (17.3)

22 (11.8)

<0.001

 T2

458 (32.8)

49 (26.2)

 T3

450 (32.3)

55 (29.4)

 T4

246 (17.6)

61 (32.6)

N stage

 N0

334 (23.9)

50 (26.7)

0.16

 N1

557 (39.9)

62 (33.2)

 N2

435 (31.2)

69 (36.9)

 N3

69 (5.0)

6 (3.2)

The PLT count was a significant difference between clinical stage and T stage, the patients with stages III–IV had higher PLT counts than in patients with stages I–II (p < 0.001) (Fig. 1). The same results were observed in patients with T3–4 than in those with T1–2 (p < 0.001) (Fig. 1). There was no significant difference between other groups.
https://static-content.springer.com/image/art%3A10.1007%2Fs13277-012-0508-y/MediaObjects/13277_2012_508_Fig1_HTML.gif
Fig. 1

a, b Platelet count according to T and clinical stages. a T stage (T1–2 vs. T3–4, p < 0.001). b Clinical stage (stages I–II vs. III–IV; p < 0.001)

PLT count and survival

The survival analysis showed that patients with thrombocytosis had a significantly shorter survival and higher distant metastatic rate than patients without thrombocytosis. No significant differences in LRFS rates were found when patients were stratified according to platelet counts. The 5-year DSS rate was 70 % in the thrombocytosis group and 78 % in the nonthrombocytosis group (p = 0.001) (Fig. 2a). Five-year MFS rate in the thrombocytosis group and nonthrombocytosis group were 81 and 88 %, respectively (p = 0.006) (Fig. 2b).
https://static-content.springer.com/image/art%3A10.1007%2Fs13277-012-0508-y/MediaObjects/13277_2012_508_Fig2_HTML.gif
Fig. 2

ae Kaplan–Meier curves showing disease-specific survival and metastases-free survival of 1,582 patients with NPC, stratified by pretreatment PLT counts. P values were determined using the log-rank test. a The 5-year DSS rates in the patients without thrombocytosis and those with thrombocytosis were 78 and 70 %, respectively, (p = 0.001). b The 5-year MFS rates in the patients without thrombocytosis and those with thrombocytosis were 88 and 81 %, respectively, (p = 0.006). c The 5-year DSS rates in stage I–II patients without thrombocytosis and those with thrombocytosis were 88 and 79 %, respectively, (p = 0.18). d The 5-year MFS rates in stage I–II patients without thrombocytosis and those with thrombocytosis were 95 and 93 %, respectively, (p = 0.43). e The 5-year DSS rates in stage III–IV patients without thrombocytosis and those with thrombocytosis were 79 and 69 %, respectively, (p = 0.006). f The 5-year MFS rates in stage III–IV patients without thrombocytosis and those with thrombocytosis were 88 and 78 %, respectively, (p = 0.002)

On subgroups analysis, no significant differences in DSS and MFS were found between thrombocytosis and nonthrombocytosis groups in patients with stages I–II (Fig. 2c, d), respectively. For patients with advanced stage (stages III–IV), significant difference was found between thrombocytosis and nonthrombocytosis groups (Fig. 2e, f).

Univariate and multivariate analyses

Table 3 summarized the univariate analysis of other relevant prognostic factors. N status, gender, chemotherapy, and high PLT count were significant predictors of distant metastasis. Age status, gender, T status, N status, chemotherapy, and high PLT count were significant factors that predicted death. Age status and T status were significant factors that predicted local regional recurrence.
Table 3

Factors associated with outcome by univariate and multivariate analyses in patients with NPC

 

Univariate

Multivariate

Outcome and variable

No. of patients (%)

P value

RR

95 % CI

P value

LRFS

 Age (≤45/>45)

817 (51.6)/765 (48.4)

0.002

1.500

1.166–1.930

0.002

 T stage (T1–2/T3–4)

770 (48.7)/812 (51.3)

<0.001

1.811

1.396–2.350

<0.001

MFS

 N stage (N0–1/N2–3)

1,003 (63.4)/579 (36.6)

<0.001

2.498

1.676–3.722

<0.001

 PLT (≤300/>300 × 109/L)

1,395 (88.2)/187 (11.8)

0.006

1.652

1.119–2.439

0.012

 Gender (female/male)

395 (25.0)/1,187 (75.0)

0.014

1.616

1.095–2.386

0.016

 Chemotherapy (yes/no)

1,064 (67.3)/518 (32.7)

<0.001

1.423

1.050–1.930

0.023

DSS

 Age (≤45/>45)

817 (51.6)/765 (48.4)

<0.001

1.913

1.528–2.396

<0.001

 Gender (female/male)

395 (25.0)/1187 (75.0)

0.013

1.381

1.050–1.816

0.021

 T stage (T1–2/T3–4)

770 (48.7)/812 (51.3)

<0.001

1.328

1.054–1.674

0.016

 N stage (N0–1/N2–3)

1,003 (63.4)/579 (36.6)

<0.001

1.471

1.174–1.843

0.001

 Chemotherapy (yes/no)

1,064 (67.3)/518 (32.7)

<0.001

1.489

1.191–1.862

0.001

 PLT (≤300/>300 × 109/L)

1,395 (88.2)/187 (11.8)

0.001

1.689

1.269–2.259

<0.001

LRFS local recurrence-free survival, MFS metastases-free survival, DSS disease-specific survival, RR relative risk, CI confidence interval

Multivariate analysis was also performed to adjust for various prognostic factors. Results were summarized in Table 3. High PLT count was found to be independently predictive of both MFS and DSS rates. Other independent factors were age, T status, N status, gender, and chemotherapy (which predicted DSS rate); age and T status (which predicted LRFS rate); and N status, chemotherapy, and gender (which predicted distant metastasis).

The efficacy of different treatment modalities was also investigated. Of those patients who received chemotherapy, 239 patients were treated with neoadjuvant chemotherapy, and 590 patients were treated with concurrent chemotherapy. The 5-year DSS, LRFS, and MFS rates in neoadjuvant and concurrent groups were 73.6 vs. 79.1 % (p < 0.001), 81.9 vs. 82.3 % (p = 0.77), and 82.3 vs. 87.2 % (p < 0.001), respectively.

Discussion

In the current study, we evaluated the relationship between PLT count and prognosis in the patients with NPC. Significant differences were observed in the PLT count according to clinical and T stages. The patients in stages III–IV had a higher PLT count compared with those in stages I–II; the similar result was observed in T stage. On the basis of the above-mentioned research, we further evaluated the effect of pretreatment PLT count on the prognosis of NPC. The original finding of the research is that PLT count for NPC appeared significantly and independently related to MFS and DSS in the analyzed data set. This relationship is apparent in a subset of patients with advanced disease (stages III–IV), but not patients with smaller tumor burden (stages I–II). To our knowledge, this is the first study to evaluate the relationship between PLT count and survival among patients with NPC.

Gonzalez et al. [13] reported on a cohort patients with lung cancer. The patients with elevated platelet counts had significantly shorter survival than patients with normal platelet counts. Similarly, in our study, we clearly demonstrated that thrombocytosis is a potential predictive factor of poor survival and high metastasis rate in patients with NPC. The association between thrombocytosis and poor oncologic outcome is not clearly defined to date. There are several possible explanations for it. According to one theory, platelets synthesize and transport several angiogenic factors such as, VEGF, platelet-derived growth factor, basic fibroblastic growth factor, epidermal growth factor, and matrix metalloproteinases, which are also known to promote tumor angiogenesis [14]. Tumor growth is dependent on the formation of new blood vessels. Platelets adhere to the tumor-related endothelium and release high concentrations of VEGF which is a potent stimulator of angiogenesis. Tumor angiogenesis is the main process responsible for the formation of new blood vessels that promote tumor growth and metastasis. Another possible explanation is that the formation of platelet-tumor cell aggregates in the circulation and facilitates immune evasion and the microvascular arrest of tumor cells at distant sites [15]. Several studies have showed that anticoagulants significantly improved survival in cancer patients [16]. More studies are required to determine the exact mechanism.

Thrombocytosis is a common finding in oncological disease, reported to be present in 10 to 57 % of patients [17]. In our study, 11.8 % of patients with NPC had thrombocytosis before treatment. The generation of platelets is a complex process, and the presence of an excess in patients with cancer is not still clear, the production of cytokines, such as interleukin 6 (IL-6) and thrombopoietin (TPO), are thought to be responsible for the increase in platelet counts [18]. TPO is an important hormone involved in differentiation of platelet precursors and is predominantly produced by the liver. TPO is produced and released into the circulation at a constant rate by the liver in normal physiological conditions. However, in thrombocytosis that can occur with malignancies, there can be upregulation of TPO production by the liver, causing increased platelet production. This study demonstrated a strong association between tumor stage and platelet count. Patients with advanced disease had higher platelet counts than those with smaller tumor burden. Similarly, in the study by Erdemir et al. [19], thrombocytosis was noted in 8 (10.81 %) of 74 patients with stage T1–2 disease and in 15 (34.09 %) of 44 patients with stage T3–4 disease (p = 0.004). In a study of patients with ovarian cancer [20], thrombocytosis was found more frequently in patients with advanced stage and a higher histologic grade. Also in colorectal cancer, an association was seen between the increase in platelets and the more advanced stages of the tumor disease [21].

The retrospective characters of the study always carry a risk of uncontrolled biases. However, this is the first report on platelet count in patients with NPC and demonstrates its clinical significance.

As the presence of thrombocytosis in some solid tumors has been associated with poor prognosis, there is increasing interest in the potential role of antithrombotic agents in the management of cancer patients. However, clinical studies have been limited and have resulted in inconsistent conclusions. Some studies suggest that antithrombotic may inhibit cancer growth and metastasis. Choe et al. [22] reported that antithrombotic was associated with an improvement in biochemical control in patients with prostate cancer who received RT with curative intent. The effect was most prominent in patients with high-risk disease. A meta-analysis of 11 studies demonstrated a significant reduction in overall mortality with anticoagulant therapy in cancer patients [23]. Phillips et al. [24] have recently demonstrated that low molecular weight heparin administration significantly increases tumor chemoresponsiveness. However, the use of antithrombotic agents in patients with cancer has been controversial. van Doormaal et al. [25] did a multicenter, randomized, open-label trial in 503 patients with advanced malignancy; a total of 244 patients were allocated to nadroparin, and 259 were allocated to the control group. A median survival of 13.1 months was observed in the nadroparin recipients compared with 11.9 months in the no-treatment arm (hazard ratio, 0.94; 95 % CI, 0.75 to 1.18, adjusted for cancer type). In addition, nadroparin had no effect on time to disease progression. These studies suggest the complicated and manifold molecular mechanisms in cancer–platelet interactions. Future studies designed to confirm the antitumor effects of antithrombotic agents and explore the pathophysiological mechanisms are awaited.

In conclusion, in the present study, we have confirmed that platelet count was not only an independent prognostic factor of poor DSS but also as an independent predictor of poor MFS in patients with NPC. Since platelet count is easily measured at low cost, it may be a useful predictor of prognosis in clinical practice. Further studies are needed to assess molecular mechanisms in cancer–platelet interactions.

Conflicts of interest

None.

Copyright information

© International Society of Oncology and BioMarkers (ISOBM) 2012