Strahlentherapie und Onkologie

, Volume 192, Issue 1, pp 47–54 | Cite as

Prognostic value of CXCL12 and CXCR4 in inoperable head and neck squamous cell carcinoma

  • Margret Rave-Fränk
  • Narges Tehrany
  • Julia Kitz
  • Martin Leu
  • Hanne Elisabeth Weber
  • Peter Burfeind
  • Henning Schliephake
  • Martin Canis
  • Tim Beissbarth
  • Holger Michael Reichardt
  • Hendrik Andreas Wolff
Original Article

Abstract

Objective

The chemokine CXCL12 and its receptor CXCR4 can affect tumor growth, recurrence, and metastasis. We tested the hypothesis that the CXCL12 and CXCR4 expression influences the prognosis of patients with inoperable head and neck cancer treated with definite radiotherapy or chemoradiotherapy.

Methods

Formalin-fixed paraffin-embedded pretreatment tumor tissue from 233 patients with known HPV/p16INK4A status was analyzed. CXCL12 and CXCR4 expressions were correlated with pretreatment parameters and survival data by univariate and multivariate Cox regression.

Results

CXCL12 was expressed in 43.3 % and CXCR4 in 66.1 % of the samples and both were correlated with HPV/p16INK4A positivity. A high CXCL12 expression was associated with increased overall survival (p = 0.036), while a high CXCR4 expression was associated with decreased metastasis-free survival (p = 0.034).

Conclusion

A high CXCR4 expression could be regarded as a negative prognostic factor in head and neck cancer because it may foster metastatic spread. This may recommend CXCR4 as therapeutic target for combating head and neck cancer metastasis.

Keywords

Head and neck cancer Squamous cell cancer Chemokines Radio(chemo)therapy Prognosis 

Prognostischer Stellenwert von CXCL12 und CXCR4 bei inoperablen Kopf-Hals-Tumoren

Zusammenfassung

Hintergrund

Das Chemokin CXCL12 und sein Rezeptor CXCR4 beeinflussen Tumorwachstum, Auftreten von Rezidiven und Metastasierung. Es wurde die Hypothese geprüft, dass ein Zusammenhang der CXCL12- und CXCR4-Expression mit der Prognose von Patienten bestehe, die wegen eines inoperablen Kopf-Hals-Tumors eine primäre Radio- oder Radiochemotherapie erhielten. Dabei wurde auch der HPV-Status der Patienten berücksichtigt.

Methodik

Formalinfixierte Proben aus unbehandelten Tumoren von 233 Patienten mit bekanntem HPV/p16INK4A-Status wurden ausgewertet. Die CXCL12- und CXCR4-Expression wurde mit klinischen Parametern und Überlebensdaten mittels uni- und multivariater Cox Regression analysiert.

Ergebnisse

CXCL12 wurde von 43,3 %, CXCR4 von 66,1 % der Tumoren exprimiert, und beide Marker korrelierten mit einer HPV/p16INK4A-Expression. Eine hohe CXCL12-Expression war mit einem verbesserten Gesamtüberleben (p = 0,036), eine hohe CXCR4-Expression mit einem reduzierten fernmetastasenfreien Überleben (p = 0,034) assoziiert.

Schlussfolgerung

Eine hohe CXCR4-Expression erscheint auch für Patienten mit inoperablen Kopf-Hals-Tumoren als negativer prognostischer Faktor, weil es Metastasierungsprozesse fördert. Eine gezielte Unterdrückung der CXCL12-/CXCR4-Signalwege kann eine Möglichkeit zur Reduktion der Metastasierung sein.

Schlüsselwörter

Kopf-Hals-Plattenepithelkarzinom Plattenepithelkarzinome Chemokin Radio(chemo)therapie Prognose 

Despite the use of surgery, chemotherapy, and radiotherapy, the 5-year survival rates for head and neck squamous cell carcinoma (HNSCC) are below 50 %. Moreover, human papilloma virus (HPV)-related tumors are more responsive to therapy than tumors related to the traditional risk factors smoking and alcohol consumption [1]. Although HNSCC is often regarded as a locoregional disease, in the end distant metastases determine the patient’s prognosis [2].

The tumor microenvironment plays a decisive role in cancer development and treatment outcome. The chemokine CXCL12 (SDF-1) and its main receptor CXCR4 have been identified as one of the key players because they affect tumor growth, tumor recurrence, metastasis, vascular formation, and therapy resistance [3, 4]. To date an interaction between CXCR4/CXCL12 and HPV status has not been clarified.

Clinically, the role of the CXCL12/CXCR4 axis in HNSCC has been investigated in some smaller studies, and for the subgroup of oral squamous cell carcinoma (OSCC) a correlation between CXCR4 expression and overall survival (OS) [5, 6, 7] as well as a correlation between CXCR4 expression and lymph node or distant metastasis was described [5, 7, 8, 9, 10]. The expression of CXCL12 was found to be higher in metastatic lymph nodes than in the primary tumor [9], and the intratumor CXCL12 level correlated with the OS [11].

The present study focused on the expression and prognostic value of CXCL12 and CXCR4 in a well-defined cohort of 233 patients with inoperable HNSCC treated with definite radiotherapy (RT) or chemoradiotherapy (CRT).

Patients and methods

Study cohort

A search of clinical records and pathology archives identified 233 patients with primary inoperable HNSCC without distant metastasis at diagnosis who were treated at a single institution. We included all patients with untreated, pathologically confirmed HNSCC with Union for International Cancer Control(UICC) stages II, III, or IV of the oral cavity, oropharynx, hypopharynx, or larynx. Patients with metastatic disease or previously treated for another cancer were excluded.

The patients’ HPV status was determined by PCR analysis and by the surrogate parameter p16INK4A [12]. Formalin-fixed paraffin-embedded (FFPE) pretreatment tumor tissue was available for all 233 patients. The clinical data of 183 patients have already been analyzed and were published in 2010 [13], and they were updated for the present investigation. The local Ethics Committee approved the use of specimens and clinicopathological data. Patient characteristics are summarized in Table 1.
Table 1

Relation between clinicopathological characteristics of 233 patients with inoperable HNSCC and the expression of CXCR4 and CXCL12/SDF-1

Characteristics

Number of patients (%)

p

Number of patients (%)

p

 

Total

CXCR4 expression

 

CXCL12/SDF-1 expression

 
  

Low ≤ 10

Median > 10–70

High > 70

 

Negative 0

Positive > 0

 

Total

n= 233

79 (33.9)

79 (33.9)

75 (32.2)

 

132 (56.7)

97 (41.6)

 

Age

    

0.0035

  

0.90

<Median

116 (49.7)

53 (45.7)

31 (26.7)

32 (27.6)

 

64 (56.1)

50 (43.9)

 

>Median

117 (50.2)

26 (22.2)

48 (41.0)

43 (36.8)

 

68 (59.1)

47 (40.9)

 

Gender

    

0.65

  

0.83

Male

197 (84.7)

67 (34.0)

69 (35.0)

61 (31.0)

 

110 (57.0)

83 (43.0)

 

Female

36 (15.3)

12 (33.3)

10 (27.8)

14 (38.9)

 

22 (61.1)

14 (38.9)

 

Tumor localization

    

0.76

  

< 0.001

Oropharynx

99 (42.5)

36 (36.4)

33 (33.3)

30 (30.3)

 

46 (48.4)

49 (51.6)

 

Hypopharynx

44 (18.9)

13 (29.5)

17 (38.6)

14 (31.8)

 

18 (40.9)

26 (59.1)

 

Larynx

27 (11.6)

10 (37.0)

10 (37.0)

7 (25.9)

 

15 (55.6)

12 (44.4)

 

Oral cavity

63 (27.0)

20 (31.7)

19 (30.2)

24 (38.1)

 

53 (84.1)

10 (15.9)

 

Histological grading

    

0.33

  

0.075

G1

11 (4.7)

3 (27.3)

3 (27.3)

5 (45.5)

 

10 (90.9)

1 (9.1)

 

G2

187 (80.3)

62 (33.2)

63 (33.7)

62 (33.2)

 

104 (56.8)

79 (43.2)

 

G3

35 (15.0)

14 (40.0)

13 (37.1)

8 (22.9)

 

18 (51.4)

17 (48.6)

 

T status

    

0.88

  

0.51

1

7 (3.0)

3 (42.9)

2 (28.6)

2 (28.6)

 

3 (42.9)

4 (57.1)

 

2

17 (7.3)

3 (17.6)

10 (58.8)

4 (58.8)

 

12 (70.6)

5 (29.4)

 

3

39 (16.7)

15 (38.5)

14 (35.9)

10 (35.9)

 

22 (56.4)

17 (43.6)

 

4

170 (73.0)

58 (34.1)

53 (31.2)

59 (31.2)

 

95 (57.2)

71 (42.8)

 

N status

    

0.78

  

0.025

0

35 (15.0)

10 (28.6)

15 (42.9)

10 (28.6)

 

26 (74.3)

9 (25.7)

 

1

27 (11.6)

11 (40.7)

8 (29.6)

8 (29.6)

 

19 (70.4)

8 (29.6)

 

2

149 (64)

48 (32.2)

51 (34.2)

50 (33.6)

 

76 (52.4)

69 (46.6)

 

3

22 (9.4)

10 (45.5)

5 (22.7)

7 (31.8)

 

11 (50.0)

11 (50.0)

 

UICC stage

    

0.53

  

0.30

II

7 (3.0)

1 (14.3)

4 (57.1)

2 (28.6)

 

4 (57.1)

3 (42.9)

 

III

16 (6.9)

4 (25.0)

6 (37.5)

6 (37.5)

 

12 (75.0)

4 (25.0)

 

IV A/B

189/21 (90.1)

74 (35.2)

69 (32.9)

67 (31.9)

 

116 (56.1)

90 (43.9)

 

RT schedule

    

0.001

  

0.53

Classic

183 (78.5)

73 (39.9)

59 (32.2)

51 (27.9)

 

104 (58.1)

75 (41.9)

 

Intensity-modulated

50 (21.5)

6 (12.0)

20 (40.0)

24 (48.0)

 

28 (56.0)

22 (44.0)

 

Chemotherapy

    

0.44

  

0.023

No

62 (26.6)

18 (29.0)

26 (41.9)

18 (29.0)

 

42 (68.9)

19 (31.1)

 

Yes

171 (73.4)

61 (35.7)

53 (31.0)

53 (31.0)

 

90 (53,6)

78 (46.4)

 

Locoregional recurrence

    

0.42

  

0.87

Events (n)

66

21

18

27

 

33

32

 

60-Month survival

0.60

0.64

0.69

0.46

 

0.55

0.64

 

120-Month survival

0.57

0.58

0.69

0.46

 

0.55

0.53

 

Distant metastases

    

0.034

  

0.52

Events (n)

29

7

5

17

 

17

12

 

60-Month survival

0.81

0.87

0.88

0.69

 

0.79

0.81

 

120-Month survival

0.81

0.87

0.88

0.69

 

0.79

0.81

 

DFS

    

0.057

  

0.89

Events (n)

81

23

22

36

 

44

36

 

60-Month survival

0.51

0.61

0.59

0.33

 

0.46

0.55

 

120-Month survival

0.48

0.55

0.59

0.33

 

0.46

0.50

 

OS

    

0.32

  

0.036

Events (n)

185

68

62

55

 

110

72

 

60-Month survival

0.22

0.17

0.25

0.24

 

0.17

0.26

 

120-Month survival

0.15

0.15

0.14

0.11

 

0.12

0.20

 

DFS disease-free survival, HNSCC head and neck squamous cell carcinoma, OS overall survival, UICC Union for International Cancer Control. Bold values indicate a statistically significant association (p < 0.05).

Treatment modalities

Patients were treated according to the respective clinical and technical standards. For 138 patients, from June 1994 to November 1999 normofractionated definite RT (2 Gy/day, 5 times/week) was delivered as parallel-opposed lateral portals. For 45 patients, from December 1999 to October 2008, normofractionated (2 Gy/day, 5 times/week) 3D conformal external-beam RT was given, the total dose being 70 Gy in each case [14]. For 50 patients, from November 2008 to November 2011, an integrated intensity-modulated radiotherapy (IMRT) with single fractions of 2.2 Gy to the primary tumor and involved lymph nodes up to 66 Gy and single fractions of 1.8 Gy to the drainage sites on both sides of the neck up to 54 Gy was applied daily (5 times/week) [12]. RT was supplemented by a concomitant chemotherapy for 171 patients, which either consisted of 5-fluorouracil plus mitomycin C or of cisplatin only.

Patient follow-up

After therapy, remission was evaluated by means of a clinical ear–nose–throat examination and contrast-enhanced computed tomography (CT). Complete remission was defined as the complete regression of all tumor manifestations. Afterward, patients underwent quarterly clinical ear–nose–throat examination, chest radiography, or CT of the head and neck, if necessary. Biopsy specimens were taken from suspect findings to receive histologic confirmation of tumor (re)growth.

Immunohistochemistry

Immunohistochemical staining of CXCR4 and CXCL12 was performed on FFPE tissue samples from pretreatment tumor biopsies. A standardized immunohistochemical staining technique was performed using rabbit-anti-CXCR4 (ab2074; dilution 1:500; Abcam, Cambridge, UK) and human/mouse-anti-CXCL12 (clone #79018; dilution 1:50; R&D Systems, Abingdon, UK) antibodies on a Ventana BenchMark XT immunostainer (Ventana, Tucson, AZ). Heat epitope retrieval using the immunostainer was performed for 60 min at 100 °C. The antibodies were incubated at 37 °C for 32 min. The staining reaction was visualized by means of horseradish peroxidase with the ultraView Universal DAB Detection Kit (Ventana Medical Systems) and hematoxylin solution (Gill 3, Sigma Aldrich, Munich, Germany) for counterstaining. Negative control slides in the absence of primary antibodies and positive controls with FFPE human cell lines of known CXCR4 and CXCL12 reactivity were included for each staining. Immunoreactivity was evaluated taking into account the fraction of positive cells and the intensity of staining. The staining intensity was scored as: 0 (absent), 1 + (weak), 2 + (moderate), and 3 + (intense). The individual weighted labeling score for CXCR4 and CXCL12 results from the addition of the products of the percentage of positive tumor cells multiplied by their staining intensity, thus scores from 0 (no positive tumor cell) to 300 (100 % intensely stained tumor cells) could be found. To assure the best possible evaluation, two blinded, independent investigators without knowledge of the clinicopathological data performed the scoring (N.T., J.K.). The individual labeling scores served for correlation analyses with clinicopathological parameters and patient survival.

Statistical analysis

Survival times were calculated from the day of histologic diagnosis until the end of study. The Kaplan–Meier method was used to estimate OS, disease-free survival (DFS), locoregional control rates (LRC), and distant metastasis-free survival (DMFS). LRC was defined as the absence of local and/or regional recurrence or progression. For the CXCR4 expression, a trisection of the data was performed. Patient characteristics were analyzed by Fisher’s exact test.

The effect of CXCR4 and CXCL12 expression on any survival parameter was analyzed by means of univariate Cox proportional hazard regressions. Furthermore, an additional multivariate analysis (multivariate Cox regression) was performed to test that the association between marker expression and survival was independent of other possible prognostic factors or factors that may influence treatment outcome, which may bias the univariate analysis. Each statistical test was performed with a significance level of α = 5 %. Statistical analyses were performed with free software (R, version 2.8; http://www.r-project.org).

Results

The median duration of follow-up was 83 months (range, 4–217 months). At the end of the analysis, 185 of 233 patients (79.4 %) had died and collectively the 5- and 10-year OS rates were 22 and 15 %, respectively. Locoregional relapse occurred in 66 of 233 patients (28.3 %), and distant metastases occurred in 29 of 233 patients (12.5 %) during follow-up.

Expression frequencies of CXCR4 and CXCL12/SDF-1

CXCR4 and CXCL12 expressions could be analyzed for 233 and 229 patients, respectively (Fig. 1). The scores for CXCR4 were from 0 to 210. Since no expression or a very low expression with a score of ≤ 10 was detected in about one third of the patients (n = 79, 33.9 %), a trisection of the data was performed. Consequently, a score between 15 and 70 was defined as medium expression (n = 79 patients, 33.9 %) and a score of > 70 was defined as a high expression (n = 75 patients, 32.2 %). Expression of CXCL12 was less frequent than in the case of CXCR4, with 132 patients (56.7 %) being negative. Here, the following analyses differentiated only between negative and positive samples. We found no correlation between the expression of CXCR4 and CXCL12 (Kendall’s tau = − 0.002, p = 0.975).
Fig. 1

Examples of tumors with positive CXCR4 (left) or positive CXCL12 (right) staining. (×40)

Associations with pretreatment parameters

For CXCR4 expression, correlation analyses showed a lack of significant association with the pretreatment parameters gender, tumor localization, T and N status, or UICC. By contrast, there was a significant increase of CXCR4 positivity for patients whose age was higher than the median (see Table 1). In the case of CXCL12, the pretreatment parameters tumor localization (84.1 % negative for oral cavity) and N status (> 70 % negative for N0, N1) were significantly associated with an existing expression. While tumors located either in the oropharynx, the hypopharynx, or the larynx showed nearly similar proportions of CXCL12 positive and negative samples, tumors of the oral cavity were rarely positive for CXCL12 (84.1 % negative vs. 15.9 % positive, see Table 1). Moreover, we noticed a significant increase of CXCL12 positivity with increasing N status (see Table 1).

Correlations with survival data

When analyzing the patients’ survival we found no significant correlations between CXCR4 expression and OS, DFS, or local recurrence-free survival (LRFS). However, a high expression of CXCR4 was significantly associated with a reduced DMFS (p = 0.034; Fig. 2a). The relationship between CXCR4 expression and the patients’ outcome is also evident from the decreased DFS for patients with high CXCR4 expression, albeit at borderline significance (p = 0.057; Fig. 2b, Table 1). We further tested whether commonly used predictors for distant metastasis such as the primary tumor site, T and N status, or histological grading correlated with DMFS, but we found no significant associations (Table 2). CXCL12 expression neither correlated with DFS nor with LRFS or DMFS. Noteworthy, a high CXCL12 expression was associated with increased OS (p = 0.036), the survival rates being 17 vs. 26 %, and 12 vs. 20 % at 5 years and 10 years, respectively (Fig. 3).
Fig. 2

Distant metastasis-free survival (a) and disease-free survival (b) according to low, medium, and high CXCR4 expression in pretreatment biopsies of patients with HNSCC treated with radiotherapy or radiochemotherapy

Fig. 3

Overall survival according to positive vs. negative staining for CXCL12 in tumor cells of pretreatment biopsies of patients with HNSCC treated with radiotherapy or radiochemotherapy

Table 2

Association of tumor localization, T status, N status, and histological grading with survival parameters (p values)

 

DFS

OS

LRFS

DMFS

Tumor localization

0.197

0.037

0.029

0.920

T status

0.174

0.705

0.126

0.781

N status

0.018

0.413

0.007

0.268

Histological grading

0.311

0.133

0.193

0.976

DFS disease-free survival, OS overall survival, LRFS local recurrence-free survival, DMFS distant metastasis-free survival. Bold values indicate a statistically significant association (p < 0.05).

Correlation with p16INK4A expression

The patients’ HPV status had already been determined by PCR analysis and by the widely accepted surrogate parameter p16INK4A [12]. Figure 4 shows that a higher CXCR4 expression as well as CXCL12 positivity is significantly correlated with a positive staining for p16INK4A.
Fig. 4

Box plots summarizing immunohistological scores for CXCR4 and CXCL12 stratified by p16INK4A positivity

Multivariate analysis

As shown in Table 1, CXCR4 expression was unevenly distributed between the RT schedules (p = 0.01), and CXCL12 expression was less frequent in patients without chemotherapy (p = 0.023). Furthermore, the degree of CXCR4 expression and positivity for CXCL12 was significantly correlated with positive staining for p16INK4A (Fig. 4). Therefore, survival data were also tested in a multivariate model. DMFS was found to be dependent on CXCR4 expression, RT schedule, and p16INK4A positivity. OS was found to be dependent on CXCL12 positivity, RT schedule, application of chemotherapy, and p16INK4A positivity. In the multivariate setting, a high CXCR4 score remained significantly correlated with a reduced DMFS (p = 0.041, Table 3), while OS was only associated with p16INK4A positivity (p = 0.034, Table 4). The use of IMRT also remained related to an increased OS, albeit at borderline significance (p = 0.058, Table 4).
Table 3

Hazard ratios and p values for the association of CXCR4 expression, RT technique (intensity-modulated vs. classic), and p16INK4A expression with distant metastasis-free survival. Bold values indicate a statistically significant association (p < 0.05).

Variable

Hazard ratio

Confidence interval

p

CXCR4

1.007

1.000–1.014

0.041

RT schedule

1.488

0.674–3.284

0.325

P16INK4Apositivity

0.999

0.996–1.003

0.751

Table 4

Hazard ratios and p values for the association of CXCL12 expression, RT technique (intensity-modulated vs. classic), chemotherapy (yes or no), and p16INK4A expression with overall survival. Bold values indicate a statistically significant association (p < 0.05).

Variable

Hazard ratio

Confidence interval

p

CXCL12/SDF-1

0.999

0.996–1.002

0.393

RT schedule

0.669

0.442–1.013

0.058

Chemotherapy

0.943

0.677–1.314

0.730

P16INK4Apositivity

0.998

0.997–0.999

0.034

Discussion

The CXCL12/CXCR4 axis plays multiple roles in HNSCC by influencing tumor growth [15], angiogenesis [16], metastasis [6, 11, 17], and patient survival [4, 11]. In the present study, immunohistochemistry was used to analyze the expression of CXCL12 and CXCR4 with respect to the clinicopathological characteristics and survival data of 233 HNSCC patients treated with definite RT or RCT. An advantage of this retrospective study was the long follow-up time of up to 217 months (median, 83 months), thereby providing reliable survival data.

Variables that are generally reported to influence the occurrence of distant metastases are tumor localization, T status, N status, and histological grading [2]. However, none of these variables was significantly associated with DMFS in the present study. After multivariate analysis, only a high CXCR4 score remained significantly correlated with a reduced DMFS. These data are in line with the findings of Ishikawa et al. [8] and Ueda et al. [10], who reported on correlations between CXCR4 mRNA or protein level and lymph node or distant metastasis. Katayama et al. [5] studied 56 patients with OSCC and found that CXCR4 positivity was an independent factor for cause-specific death.

The underlying mechanism of increased CXCR4 expression in HNSCC and the relation to enhanced tumor metastasis and inferior patient survival are generally ascribed to diverse factors, including EMT (epithelial–mesenchymal transition) induction and activation of matrix metalloproteinases. Ou et al. suggested that the EMT-related transcription factor Twist regulates CXCR4 expression in HNSCC, which in turn might be associated with lymph node metastasis [18]. Taki et al. reported that the CXCR4 expression in oral OSCC cells was up-regulated by CXCL12 and transforming growth factor β1 in a Snail-dependent manner [19], which is also an EMT promoting transcription factor. Another important CXCR4 regulating factor is hypoxia, which is common in HNSCC, and may induce the enhanced expression of CXCR4 via activation of hypoxia-inducible factor-1 [20]. Clinical and experimental evidence for the relationship between CXCR4 expression and tumor cell migration or tumor metastasis already led to the assumption that CXCR4 targeting by monoclonal antibodies or small molecule inhibitors could become an efficient strategy for treating human cancer metastasis, and the first clinical trials are ongoing [3, 21].

To our knowledge, data on the herein described correlation between high CXCR4 expression and p16INK4A/HPV tumor positivity have not been published to date. However, patients with WHIM (warts, hypogammaglobulinemia, infections, and myelokathexis) syndrome, an autosomal dominant disease caused by a gain of function mutation in CXCR4, show increased susceptibility to HPV infections [22]. WHIM manifests as cutaneous warts, cervical dysplasia, and squamous carcinoma in women, and the incidence of HPV-related SSC of the oral cavity in two relatives with WHIM syndrome has been described [23]. The mutations present in WHIM patients probably result in enhanced CXCR4 signaling and an increase of the normal adhesion-promoting function of CXCR4, which is consistent with the increased CXCR4 expression of patients with distant metastases as described here.

Clinical studies addressing the association between CXCL12 and patient outcome are still scarce. We now observed that in a univariate setting CXCL12expression was associated with increased OS. This finding confirms the results of Clatot et al. [11], who found in a series of 71 OSCC patients that CXCL12 levels were significantly correlated with metastatic evolution and OS. Furthermore, in a gene expression study by these authors [24] higher expression levels of genes involved in CXCL12 signaling were observed in the group with better prognosis. Observations made in experimental studies support the clinical findings of better prognoses at high CXCL12 expression, too. In the case of HNSCC, it was suggested that high CXCL12 expression in the primary tumor or in the tumor microenvironment suppresses the CXCR4-driven chemotactic migration of tumor cells toward distant tissues expressing CXCL12 [4, 24]. It should be pointed out that in a multivariate setting, where the association of CXCL12 expression, RT technique (intensity-modulated vs. classic), chemotherapy (yes or no), and p16INK4A expression with OS was analyzed, only p16INK4A expression remained significantly associated with OS. Therefore, it is also conceivable that due to the strong correlation between CXCL12 and p16INK4A expression, the positive prognostic value of CXCL12 simply mirrors the well-known advantageous effect of p16INK4A expression on patient survival [25].

Conclusion

In summary, we have presented clinical data underscoring the prognostic value of CXCR4 in HNSCC patients treated with RT or RCT. The association between inferior DMFS and high CXCR4 expression is not only useful in estimating the prognosis but also recommends the CXCL12/CXCR4 axis as a novel therapeutic target for treating HNSCC metastasis.

Notes

Compliance with ethical guidelines

Conflict of interest

M. Rave-Fränk, N. Tehrany, J. Kitz, M. Leu, H.E. Weber, P. Burfeind, H. Schliephake, M. Canis, T. Beissbarth, H.M. Reichardt, and H.A. Wolff state that there are no conflicts of interest.

References

  1. 1.
    Heiduschka G, Grah A, Oberndorfer F et al (2015) Improved survival in HPV/p16-positive oropharyngeal cancer patients treated with postoperative radiotherapy. Strahlenther Onkol 191:209–216PubMedCrossRefGoogle Scholar
  2. 2.
    Takes RP, Rinaldo A, Silver CE et al (2012) Distant metastases from head and neck squamous cell carcinoma. Part I. Basic aspects. Oral Oncol 48:775–779PubMedCrossRefGoogle Scholar
  3. 3.
    Domanska UM, Kruizinga RC, Nagengast WB et al (2013) A review on CXCR4/CXCL12 axis in oncology: no place to hide. Eur J Cancer 49:219–230PubMedCrossRefGoogle Scholar
  4. 4.
    Albert S, Riveiro ME, Halimi C et al (2013) Focus on the role of the CXCL12/CXCR4 chemokine axis in head and neck squamous cell carcinoma. Head Neck 35:1819–1828PubMedCrossRefGoogle Scholar
  5. 5.
    Katayama A, Ogino T, Bandoh N et al (2005) Expression of CXCR4 and its down-regulation by IFN-gamma in head and neck squamous cell carcinoma. Clin Cancer Res 11:2937–2946PubMedCrossRefGoogle Scholar
  6. 6.
    Uchida D, Onoue T, Tomizuka Y et al (2007) Involvement of an autocrine stromal cell derived factor-1/CXCR4 system on the distant metastasis of human oral squamous cell carcinoma. Mol Cancer Res 5:685–694PubMedCrossRefGoogle Scholar
  7. 7.
    Salcedo R, Wasserman K, Young HA et al (1999) Vascular endothelial growth factor and basic fibroblast growth factor induce expression of CXCR4 on human endothelial cells: in vivo neovascularization induced by stromal-derived factor-1alpha. Am J Pathol 154:1125–1135PubMedPubMedCentralCrossRefGoogle Scholar
  8. 8.
    Ishikawa T, Nakashiro K, Hara S et al (2006) CXCR4 expression is associated with lymph-node metastasis of oral squamous cell carcinoma. Int J Oncol 28:61–66PubMedGoogle Scholar
  9. 9.
    Keeley EC, Mehrad B, Strieter RM (2010) CXC chemokines in cancer angiogenesis and metastases. Adv Cancer Res 106:91–111PubMedPubMedCentralCrossRefGoogle Scholar
  10. 10.
    Ueda M, Shimada T, Goto Y et al (2010) Expression of CC-chemokine receptor 7 (CCR7) and CXC-chemokine receptor 4 (CXCR4) in head and neck squamous cell carcinoma. Auris Nasus Larynx 37:488–495PubMedCrossRefGoogle Scholar
  11. 11.
    Clatot F, Picquenot JM, Choussy O et al (2011) Intratumoural level of SDF-1 correlates with survival in head and neck squamous cell carcinoma. Oral Oncol 47:1062–1068PubMedCrossRefGoogle Scholar
  12. 12.
    Tehrany N, Kitz J, Rave-Fränk M et al (2015) High-grade acute organ toxicity and p16 expression as positive prognostic factors in primary radio(chemo)therapy for patients with head and neck squamous cell carcinoma. Strahlenther Onkol 191:566–572PubMedCrossRefGoogle Scholar
  13. 13.
    Wolff HA, Bosch J, Jung K et al (2010) High-grade acute organ toxicity as positive prognostic factor in primary radio(chemo)therapy for locally advanced, inoperable head and neck cancer. Strahlenther Onkol 186:262–268PubMedCrossRefGoogle Scholar
  14. 14.
    Wolff HA, Daldrup B, Jung K et al (2011) High-grade acute organ toxicity as positive prognostic factor in adjuvant radiation and chemotherapy for locally advanced head and neck cancer. Radiology 258:864–871PubMedCrossRefGoogle Scholar
  15. 15.
    Meng X, Wuyi L, Yuhong X et al (2010) Expression of CXCR4 in oral squamous cell carcinoma: correlations with clinicopathology and pivotal role of proliferation. J Oral Pathol Med 39:63–68PubMedCrossRefGoogle Scholar
  16. 16.
    Li KC, Huang YH, Ho CY et al (2012) The role of IL-8 in the SDF-1alpha/CXCR4-induced angiogenesis of laryngeal and hypopharyngeal squamous cell carcinoma. Oral Oncol 48:507–515PubMedCrossRefGoogle Scholar
  17. 17.
    Uchida D, Onoue T, Kuribayashi N et al (2011) Blockade of CXCR4 in oral squamous cell carcinoma inhibits lymph node metastases. Eur J Cancer 47:452–459PubMedCrossRefGoogle Scholar
  18. 18.
    Ou DL, Chien HF, Chen CL et al (2008) Role of Twist in head and neck carcinoma with lymph node metastasis. Anticancer Res 28:1355–1359PubMedGoogle Scholar
  19. 19.
    Taki M, Higashikawa K, Yoneda S et al (2008) Up-regulation of stromal cell-derived factor-1alpha and its receptor CXCR4 expression accompanied with epithelial-mesenchymal transition in human oral squamous cell carcinoma. Oncol Rep 19:993–998PubMedGoogle Scholar
  20. 20.
    Ishikawa T, Nakashiro K, Klosek SK et al (2009) Hypoxia enhances CXCR4 expression by activating HIF-1 in oral squamous cell carcinoma. Oncol Rep 21:707–712PubMedGoogle Scholar
  21. 21.
    Ramsey DM, McAlpine SR (2013) Halting metastasis through CXCR4 inhibition. Bioorg Med Chem Lett 23:20–25PubMedCrossRefGoogle Scholar
  22. 22.
    McDermott DH, Liu Q, Velez D et al (2014) A phase 1 clinical trial of long-term, low-dose treatment of WHIM syndrome with the CXCR4 antagonist plerixafor. Blood 123:2308–2316PubMedPubMedCentralCrossRefGoogle Scholar
  23. 23.
    Cipriani NA, Blair E, Taxy JB (2010) WHIM syndrome and oral squamous cell carcinoma. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 109:105–108PubMedCrossRefGoogle Scholar
  24. 24.
    Clatot F, Gouerant S, Mareschal S et al (2014) The gene expression profile of inflammatory, hypoxic and metabolic genes predicts the metastatic spread of human head and neck squamous cell carcinoma. Oral Oncol 50:200–207PubMedCrossRefGoogle Scholar
  25. 25.
    Rainsbury JW, Ahmed W, Williams HK et al (2013) Prognostic biomarkers of survival in oropharyngeal squamous cell carcinoma: systematic review and meta-analysis. Head Neck 35:1048–1055PubMedCrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2015

Authors and Affiliations

  • Margret Rave-Fränk
    • 1
  • Narges Tehrany
    • 1
  • Julia Kitz
    • 2
  • Martin Leu
    • 1
  • Hanne Elisabeth Weber
    • 1
  • Peter Burfeind
    • 3
  • Henning Schliephake
    • 4
  • Martin Canis
    • 5
  • Tim Beissbarth
    • 6
  • Holger Michael Reichardt
    • 7
  • Hendrik Andreas Wolff
    • 1
  1. 1.Department of Radiotherapy and Radiation OncologyUniversity Medical Center GöttingenGöttingenGermany
  2. 2.Department of PathologyUniversity Medical Center GöttingenGöttingenGermany
  3. 3.Department of Human GeneticsUniversity Medical Center GöttingenGöttingenGermany
  4. 4.Department of Oral and Maxillofacial SurgeryUniversity Medical Center GöttingenGöttingenGermany
  5. 5.Department of Otorhinolaryngology, Head and Neck SurgeryUniversity Medical Center GöttingenGöttingenGermany
  6. 6.Institute of Medical StatisticsUniversity Medical Center GöttingenGöttingenGermany
  7. 7.Institute for Cellular and Molecular ImmunologyUniversity Medical Center GöttingenGöttingenGermany

Personalised recommendations