Annals of Surgical Oncology

, Volume 20, Issue 7, pp 2419–2427

Correlation of p53 Status with the Response to Chemotherapy-Based Treatment in Esophageal Cancer: A Meta-Analysis

  • Shui-Shen Zhang
  • Qing-Yuan Huang
  • Hong Yang
  • Xuan Xie
  • Kong-Jia Luo
  • Jing Wen
  • Xiao-Li Cai
  • Fu Yang
  • Yi Hu
  • Jian-Hua Fu
Thoracic Oncology

DOI: 10.1245/s10434-012-2859-4

Cite this article as:
Zhang, SS., Huang, QY., Yang, H. et al. Ann Surg Oncol (2013) 20: 2419. doi:10.1245/s10434-012-2859-4

Abstract

Background

The value of p53 status for predicting response to chemotherapy-based treatment in patients with esophageal cancer has been controversial. We conducted a meta-analysis to elucidate the correlation of p53 status with the response to chemotherapy-based treatment.

Methods

Studies were searched in PubMed, Embase, and Web of Science (up to September 2012). The p53 status and response to therapy were defined and standardized. Subgroup analyses based on the treatment and histopathology were performed to explore the usefulness of p53 status for predicting response to therapy in esophageal cancer. Sensitivity analyses were conducted by removing specific studies to assess the effects of study quality.

Results

We included 28 studies with 1497 cases in our meta-analysis. Wild-type form of p53 status (low expression of p53 protein and/or wild-type p53 gene) was associated with high response to chemotherapy-based treatment in esophageal cancer (total major response [MR]: risk ratio [RR] = 1.09, 95 % CI = 1.03–1.16, P = .003; pathological MR: RR = 1.15, 95 % CI = 1.06–1.25, P = .001; total complete response [CR]: RR = 1.08, 95 % CI = 1.00–1.17, P = .040). The similar correlation between the wild-type form p53 and response to therapy were also detected in subgroup analyses (total MR, pathological MR, and total CR in chemoradiotherapy subgroup; total MR in chemotherapy subgroup; total MR and pathological CR in esophageal squamous cell carcinoma [ESCC]). Additionally, patients with wild-type form p53 status had high pathological complete response rate to neoadjuvant chemoradiotherapy in ESCC.

Conclusions

The current meta-analysis suggested that p53 status might be a predictive biomarker for response to chemotherapy-based treatment in esophageal cancer.

Esophageal cancer is one of the most common cancers in the world, with more than 480,000 new cases and 400,000 deaths occurring annually worldwide.1 Though surgery plays a central role in the overall management of this devastating disease, its prognosis remains poor.2 Nowadays, it is clear that neoadjuvant chemoradiotherapy (NCRT) or chemotherapy (NCT) can improve the outcome of esophageal cancer.3,4 However, some studies suggested that only responders would benefit from neoadjuvant therapy, while neoadjuvant therapy could worsen the prognosis of nonresponders.5, 6, 7 Therefore, predictive markers to allow individualization of multimodality therapy would be invaluable as it would identify those individuals who will benefit from this treatment strategy in locally advanced esophageal cancer.

P53 is a candidate biomarker for predicting the response of esophageal cancer to modality treatment and has been the most studied.8 As a tumor suppressor gene, p53 is the most widely mutated gene in human carcinogenesis, with mutations occurring in at least 40 % of all esophageal tumors.9,10 While it is a critical transcription factor, it is also involved in key cellular functions such as in cell cycle regulation, apoptosis, and DNA repair.11 Thus, the status of p53 tends to associate with tumor tolerance to chemotherapy or radiation.12,13 These results have been demonstrated in vivo and in vitro.14,15

However, there is no general consensus on the issue of p53 status to predict the response of esophageal cancer to chemotherapy (CT) or chemoradiotherapy (CRT). The findings to date have yielded conflicting results. Several studies suggested that patients with p53 status (wild-type or negative expression) often had a higher major response rate to CT or CRT than those with mutation p53 or overexpression.16, 17, 18, 19, 20 However, some studies found the difference results.21, 22, 23, 24, 25, 26 Therefore, we aimed to elucidate the value of p53 status for predicting response to chemotherapy-based treatment in esophageal cancer by meta-analysis.

Methods

Search Strategy

Two reviewers independently performed a systematic literature search of the following databases: PubMed, Embase, and Web of Science (last search up to September 2012). The following search terms were used: “TP53 or p53 or p53 protein or p53 mutation or 17p13 gene,” “chemotherapy or chemoradiotherapy,” and “esophageal cancer.” All potentially eligible studies were retrieved.

Inclusion and Exclusion Criteria

Studies were included if they met all of the following criteria: (1) evaluation of p53 status for predicting the response to CT or CRT in esophageal cancer, (2) description of clinical or pathological response, (3) retrospective or prospective cohort studies, and (4) studies published in English. Reviews, letters to the editor, and articles published in books were excluded. Disagreements between reviewers were resolved by a third reviewer or by discussion and consensus.

Definitions and Standardizations

The definition and standardization for “p53” and “response to therapy” was on the basis of the study published by Pakos et al.27 For consistency, we used “p53 status” for covering both the gene and protein markers. Mutation of p53 gene increases the half-life of p53 protein, leading to nuclear accumulation of mutant TP53, which can be detected by IHC.28,29 However, p53 protein accumulation measured by IHC does not necessarily correspond to p53 gene mutation detected by RT-PCR.30 Thus, the overall analysis considered all studies, regardless of whether protein expression or gene mutation was being evaluated. Because different cutoff thresholds were used for the definition of a positive test in studies by IHC, we defined p53 protein positivity as nuclear cell stain in at least 10 % of the tumor cells, a definition followed by most studies. When different definitions were used, we accepted the cutoff closest to the 10 % level. Wild-type form of p53 status was defined as low expression of p53 protein and/or wild-type p53 gene while mutation-type form of p53 status meant high expression p53 protein and/or mutation-type p53 gene. Response was defined as grade3 or complete response (CR), grade 2 or partial response (PR), or major response (MR) (MR = grade 3 + grade 2 or MR = CR + PR), according to the guidelines for the clinical and pathologic studies on carcinoma of the esophagus by the Japanese Society of Esophageal Diseases (JSED) or RECIST (Response Evaluation Criteria in Solid Tumors) criteria.31,32 For consistency, we define the response classification in Table 1.
Table 1

Criteria for response evaluation and standard definition

Criteria

Standard definition

Minor response

Major response

Complete response

JSED13

Grade 0 + 1, viable cancer cells account for more than 1/3

Grade 2 + 3, viable cancer cells account for less than 1/3

Grade 3, no residual viable tumor cells

RECIST32

PD + SD, <30 % regression of the disease

PR + CR, ≥ 30 % regression of the disease

CR, 100 % regression of the disease

WHO

NC + PD, <50 % decrease in tumor load

PR + CR, > 50 % decrease in tumor load

CR, disappearance of all known disease

Nasierowska et al.52

PR2 + SD

PR1 + CR, single cells or small nests of cancer cells or no cancer cells

CR, no microscopic evidence of cancer cells

Becker et al.6 and Brucher et al.7

>10 % residual tumor cells

<10 % residual tumor cells

No residual tumor cells

Beardsmore et al.44

No response, absence of tumor downstaging

PR + CR, evidence of downstaging

CR, absence of tumor in resected specimen

Yang et al.19

ORT, grossly residual tumor and/or residual tumor in 2 or more tissue blocks

NRT + MRT, absence of tumor either grossly or microscopically

NRT, absence of tumor both grossly and microscopically

JSED Japanese Society for Esophageal Disease, PD progressive disease, SD stable disease, PR partial response, CR complete response, NC no change, ORT overt residual tumor, MRT minimal residual tumor, NRT no residual tumor

Data Extraction

The data extracted from each study were: author’s name, publication year, cases of patients enrolled, histopathology type of cancer, treatment, methods of detection, type of therapeutic response, p53 positive (overexpression and mutation) rate, and the response rate. Two investigators extracted data from eligible studies independently, discussed discrepancies, and reached consensus for all items.

Statistical Analysis

Meta-analysis was performed with the software STATA version 12 (StataCorp, College Station, TX). We assessed and quantified statistical heterogeneity for each pooled estimate using the I2 statistic. If heterogeneity existed, a random-effects model was adopted; otherwise, a fixed-effects model was used. Pooled analysis was performed using the Mantel-Haenszel model and reported as risk ratio (RR) with 95 % CIs. The significance of the pooled RR was determined by the z test, and P < .05 was considered statistically significant. The Begg’s funnel plot and Egger’s test were used to estimate the potential publication bias. Sensitivity analysis was conducted to reevaluate the overall results by omitting specific studies.

Results

Eligible Studies

As the search flow diagram shows (Fig. 1), 28 studies with 1497 cases in total were included in our meta-analysis.9,17, 18, 19, 20, 21, 22, 23, 24, 25, 26,33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49 The characteristics of the eligible studies are summarized in Table 2; 18 studies used CRT (including 8 NCRT), 6 studies used CT, and 2 studies consisted of both NCRT and NCT. One study with hyperthermochemoradiotherapy (HCRT) and the other study using CRT and radiotherapy were also included.48,49 The histopathology in most of the included studies was squamous cell carcinoma (SCC), while 3 studies showed both SCC and adenocarcinoma (AC) and 1 study showed AC of the esophagus and gastroesophageal junction (GEJ). A total of 24 studies used protein detection with immunohistochemistry (IHC), while only 1 used gene detection (including polymerase chain reaction-single strand conformation polymorphism [PCR-SSCP] and functional assay), and 4 used both methods. The cases of patients in all eligible studies ranged from 18 to 107. The p53 positive rate ranged from 27.5 % to 85.3 % (mean = 58.0 %, SD = 2.60 %). The total (clinical and pathological) major response rate ranged from 10.0 % to 86.4 % (mean = 54.0 %, SD = 4.20 %), and the total (clinical and pathological) CR (complete response) rate ranged from 1.61 % to 64.7 % (mean = 34.2 %, SD = 4.41 %). There were 19 studies conducted in Japan, 1 in Korea, 6 in Europe, and 2 in the United States.
Fig. 1

Flow diagram illustrating the screening and selection process

Table 2

Characteristics of studies included in the meta-analysis

Author

Year

Country

Treatment

N

Hp

Detection

p53 positive rate (%)

Type of response

Response criteria

Standard definition response

Response rate (%)

Minor response

Major response

CR

Major response

Complete response

Yamamoto et al.21

2012

Japan

NCT

37

SCC

IHC

48.6

Pathological

JSED

G 0 + 1

G 2 + 3

NR

40.5

NR

   

CT

51

SCC

IHC

56.8

Clinical

RECIST

PD + SD

CR + PR

NR

64.7

NR

Yamasaki et al.17

2010

Japan

CT

97

SCC

PCR-SSCP

47.4

Clinical

RECIST

PD + SD

CR + PR

NR

43.3

NR

      

IHC

56.7

       

Makino et al.9

2010

Japan

CRT

64

SCC

PCR-SSCP

31.3

Pathological

JSED

NR

G 3

G3

 

21.9

      

IHC

65.6

       

Fukuchi et al.23

2009

Japan

CRT

68

SCC

IHC

39.7

Clinical

JSED

NC + PD

CR + PR

CR

67.6

23.5

Nam et al.24

2008

Korea

CRT

51

SCC

IHC

27.5

Pathological

CR

NR

CR

CR

 

64.7

        

Clinical

CR

NR

CR

CR

 

49.0

Sarbia et al.25

2007

Germany

CRT

90

SCC

IHC

77.8

Clinical

WHO

NC + PD

CR + PR

NR

54.4

NR

        

Pathological

Becker et al.

Brucher

et al.

>10 % residual

tumor cells

<10 % residual

tumor cells

NR

64.4

NR

Ishida et al.49

2007

Japan

HCRT

32

SCC

IHC

53.1

Pathological

JSED

G 1 + 2

G 3

G 3

 

37.5

Sunada et al.20

2005

Japan

CRT

36

SCC

IHC

58.3

Clinical

WHO

NC + PD

CR + PR

NR

69.4

NR

Miyazaki et al.48

2005

Japan

CRT + RT

61

SCC

IHC

45.9

Clinical

JSED

G 0 + 1

G 2 + 3

NR

73.8

NR

Okumura et al.18

2005

Japan

CRT

62

SCC

IHC

43.5

Clinical

JSED

NC + PD

CR + PR

CR

71.0

1.61

        

Pathological

    

52.8

22.2

Sohda et al.26

2004

Japan

CRT

65

SCC

IHC

41.5

Clinical

JSED

NC + PD

CR + PR

CR

69.2

24.6

Heeren et al.46

2004

Netherlands

NCT

30

AC + GEJ

IHC

73.3

Clinical

WHO

NC + PD

CR + PR

NR

27.8

 

Nakamura et al.47

2004

Japan

CRT

76

SCC

IHC

57.9

Clinical

JSED

NC + PD

CR + PR

CR

72.4

21.1

Takeuchi et al.45

2003

Japan

CRT

41

SCC

IHC

56.1

Clinical

JSED

NC + PD

CR + PR

NR

70.7

NR

Kishi et al.22

2003

Japan

NCT

107

SCC

IHC

70.1

Clinical

RECIST

PD + SD

CR + PR

NR

61.7

NR

        

Pathological

JSED

G 0 + 1a

G 1b + 2

NR

27.1

NR

Beardsmore et al.44

2003

UK

NCRT + NCT

48

SCC + AC

IHC

70.8

Clinical+

pathological

Beardsmore

et al.

No response

PR + CR

CR

68.8

25

Michel et al.43

2002

France

CRT

34

SCC

IHC

63.6

Clinical+

pathological

CR

NR

CR

CR

 

60.6

      

FA

85.3

      

61.8

Shimada et al.42

2002

Japan

CRT

52

SCC

IHC

57.7

Clinical

JSED

NC + PD

CR + PR

NR

69.2

NR

Kajiyama et al.41

2002

Japan

NCT

60

SCC

IHC

58.3

Pathological

JSED

G 0 + 1

G2 + 3

NR

10.0

NR

   

NCRT

22

SCC

IHC

68.2

Pathological

JSED

G 0 + 1

G 2 + 3

NR

86.4

NR

Ito et al.40

2001

Japan

CRT

40

SCC

PCR-SSCP

60.0

Clinical

WHO

NR

CR-

CR

 

32.5

Takeno et al.39

2001

Japan

CRT

34

SCC

IHC

29.4

Pathological

JSED

G 0 + 1

G 2 + 3

G 3

41.2

8.82

Szumilo et al.38

2000

Poland

NCT

34

SCC

IHC

67.6

Pathological

Nasierowska

et al.52

PR2 + SD

PR1 + CR

NR

23.5

NR

Shimada et al.37

2000

Japan

NCT

59

SCC

IHC

44.1

Pathological

JSED

G 0 + 1a

G 1b + 2

NR

15.3

NR

Krasna et al.35

1999

USA

CRT

22

SCC + AC

IHC

81.8

Pathological

CR

NR

CR

CR

 

36.4

Samejima et al.36

1999

Japan

CRT

20

SCC

IHC

70.0

Pathological

JSED

G 0 + 1

G 2 + 3

G 3

70.0

20

Yang et al.19

1998

USA

CRT

95

SCC + AC

IHC

71.9

Pathological

Yang et al.

ORT

NRT + MRT

NRT

35.9

NR

Puglisi et al.33

1996

Italy

CRT

22

SCC

IHC

77.3

Pathological

CR

NR

CR

CR

 

54.5

Muro et al.34

1996

Japan

CRT

18

SCC

IHC

55.6

Clinical

CR

NR

CR

CR

 

50.0

N number of patients, NCRT neoadjuvant chemoradiotherapy, NCT neoadjuvant chemotherapy, CT chemotherapy, CRT chemoradiotherapy, HCRT hyperthermochemoradiotherapy, RT radiotherapy, SCC squamous cell carcinoma, AC adenocarcinoma, GEJ gastroesophageal junction, IHC immunohistochemistry, PCR-SSCP polymerase chain reaction-single strand conformation polymorphism, FA functional assay, Hp histopathology, JSED Japanese Society for Esophageal Disease, PD progressive disease, SD stable disease, PR partial response, CR complete response, NC no change, ORT overt residual tumor, MRT minimal residual tumor, NRT no residual tumor, G grade, NR no record

Correlation of p53 Status with Response to Chemotherapy-Based Treatment in Esophageal Cancer

A total of 28 studies involving 1497 esophageal cancer patients who received chemotherapy-based treatment were identified to pooled estimate of the correlation of p53 status with total MR. As Fig. 2 shows, wild-type form of p53 status was significantly correlated with total MR (RR = 1.09, 95 % CI = 1.03–1.16, P = .003). With respect to studies using both clinical and pathological response, the later data was used. We also examined the clinical response data and found similar results (RR = 1.10, 95 % CI = 1.04–1.17, P = .001). Besides, wild-type form of p53 status was significantly associated with improved pathological MR (RR = 1.15, 95 % CI = 1.06–1.25, P = .001) and total CR (RR = 1.08, 95 % CI = 1.00–1.17, P = .040) (Fig. 3). There were 7 studies identified to pooled estimated of pathological CR. Patients with wild-type form of p53 status trended to have a higher pathological complete response rate to chemotherapy-based treatment, but did not reach significance (RR = 1.11, 95 % CI = 0.99–1.24, P = .069). In all above pooled estimates, no significant heterogeneity was detected (Supplement Table 1).
Fig. 2

Forest plots of RR were estimated for association between p53 status and total major response (MR) to chemotherapy-based treatment in patients with esophageal cancer

Fig. 3

Forest plots of RR were estimated for association between p53 status and response to chemotherapy-based treatment in patients with esophageal cancer a for pathological major response (MR) and b for total complete response (CR)

Subgroup Analysis

Among all studies using chemotherapy-based treatment, the results of the CRT and CT were estimated separately. Several studies using CT were included to examine the relation between p53 status and response to chemotherapy. There was significant correlation of wild-type form of p53 status with total MR (RR = 1.11, 95 % CI = 1.01–1.21, P = .021), but not with pathological MR (RR = 1.07, 95 % CI = 0.97–1.19, P = .159). For studies using both clinical and pathological response, the later data was used. We also examined the clinical response data and found similar results (RR = 1.14, 95 % CI = 1.04–1.25, P = .007). With respect to CRT, 22 studies contributed data on total MR setting. Wild-type form of p53 status was associated with improved total MR (RR = 1.09, 95 % CI = 1.01–1.18, P = .023). For studies using both clinical and pathological response, the pathological response data was used. We also examined the clinical response data and found similar results (RR = 1.09, 95 % CI = 1.01–1.17, P = .026). Also, significant correlation of wild-type form of p53 status with pathological MR (RR = 1.12, 95 % CI = 1.01–1.25, P = .042) and total CR (RR = 1.07, 95 % CI = 1.00–1.16, P = .048) to CRT was also detected. However, patients with wild-type form of p53 status trended to have a high pathological CR rate to CRT treatment, but did not reach significance (RR = 1.11, 95 % CI = 0.99–1.24, P = .069).

Because the histopathology in most studies was squamous cell carcinoma, we therefore separately examined the correlation of p53 status with the response to chemotherapy-based in patients with esophageal squamous cell carcinoma (ESCC). Wild-type form of p53 status was associated with the total MR (RR = 1.01, 95 % CI = 1.02–1.16, P = .007) and pathological CR (RR = 1.13, 95 % CI = 1.00–1.26, P = .043), but not with pathological MR (RR = 1.07, 95 % CI = 0.98–1.16, P = .111) or total CR (RR = 1.07, 95 % CI = 0.99–1.16, P = .087). Examining whether p53 status was associated with the complete response to NCRT in ESCC, 6 studies were identified. Patients with wild-type form p53 status had high pathological complete response rate (RR = 1.13, 95 % CI = 1.00–1.26, P = .043) to NCRT (Supplement Fig. 1). In all above pooled estimates, no significance of heterogeneity was detected (Supplement Table 1).

Publication Bias and Sensitivity Analysis

Begg’s funnel plot and Egger’s test were used to estimate the potential publication bias of the included literature. The funnel plots showed no evidence of obvious asymmetry (Supplement Fig. 2), and Egger’s test indicated no significance of publication bias (P > .05). Moreover, sensitivity analysis was carried out to assess the influence of individual studies on the summary effect. Removal of 1 study used both CRT and radiotherapy, and 1 study with HCRT, did not alter the overall results (Supplement Table 1).

Discussion

P53, as a surrogate biomarker, has been studied mostly to predict the response to chemotherapy-based treatment. However, most of the studies yielded conflicting results and this issue has always been controversial. The reasons can be summarized: lack of large-scale clinical studies, the main detection techniques lacking high sensitivity and specificity, different measurement of p53 status, no standard evaluation of response to therapy, and no standardized regimes of chemotherapy-based therapy, specifically CRT. Thus, our current meta-analysis with a large-scale population was the best way and the first time the correlation of p53 status with the response to chemotherapy-based treatment in esophageal cancer was systematically elucidated.

The results of our meta-analysis indicated the wild-type form of p53 status may predict the high response rate to chemotherapy-based treatment in patients with esophageal cancer. Wild-type form of p53 status was correlated with improved total MR, CR, and pathological MR. With respect to the correlation with increased pathological CR, there was a marginal trend with nonsignificance. The sensitivity analysis had no significant effect on the overall results. Previous studies described a higher response rate in patients with p53-negative, while others found no significant discrepancies in response between p53 status.9,17,20, 21, 22, 23, 24, 25, 26,37, 38, 39, 40,42, 43, 44, 45,47, 48, 49 These conflicts were attributed to 3 reasons. The first reason was the potentially controversial nature of staining interpretation. Interpretation of p53 detection by IHC analysis was based on the premise that wild-type p53 status has a short half-life and was not detectable, whereas mutant p53 is more stable and detectable.28,29 P53 protein accumulation measured by IHC does not necessarily correspond to p53 gene mutation detected by RT-PCR.30 This may be explained by a nonmutational mechanism for p53 protein accumulation, such as inactivation of the function for p53 protein degradation.50 Secondly, the evaluation of response to therapy among those studies was of a great difference and variety. Some studies used the criteria of JSED, while some used the RECIST criteria. A few studies even had their own criteria.19,44 Standardization was therefore of great significance for obtaining an accurate assessment of the clinical significance of p53 status and response of the therapy. We used the definition and standard in our meta-analysis in order to reduce the heterogeneity among the included studies, which made our results more reliable and representative. Thirdly, different regimens of therapy were used among these studies (Supplement Table 2). The dose of chemotherapy or radiation and courses of therapy were variable.

On stratification analysis according to the treatment, we found the similar result that the relation remained for subgroup of total MR, CR, and pathological MR in patients receiving CRT, except for pathological CR. Additionally, patients with wild-type form of p53 status may have a high total major response (MR) rate to chemotherapy (including NCT), but did not have an increased pathological major response (MR) rate. Further analysis based on the histopathology revealed that wild-type form of p53 status was associated with relevant increases in total MR and pathological CR to chemotherapy-based treatment in patients with esophageal squamous cell carcinoma (ESCC). Although wild-type form of p53 status trended to be correlated with pathological MR and total CR, there was no significance with a marginal P value. We also found that wild-type form of p53 status was associated with high pathological CR rate to NCRT in patients with ESCC, while some authors reported that only patients with pathological complete response to NCRT could have a long-term survival benefit in ESCC.51 Therefore, p53 status may act as a clinical surrogate biomarker to identify those who really benefit from NCRT and prevent those who did not respond to NCRT from delaying the appropriate therapy procedure as well as getting rid of the toxicities of anticancer drugs and radiation therapy.

We acknowledge that our meta-analysis suffered from several limitations despite our attempts to perform a comprehensive analysis. First, the meta-analysis may have been influenced by publication bias; we limited the search to only the studies published in English, and we did not search abstract reported in major scientific meetings and abstract books, which may have introduced publication bias to meta-analysis. Although we tried to identify all relevant data, some missing data were unavoidable. Second, cutoff values of p53 for both overexpression by IHC and gene mutation by PCR-SSCP were variable in each study, which might lead to inconsistent results across the included studies. Third, 2 studies with special treatment such as HCRT and RT were also included in our meta-analysis, although sensitivity analysis by removal of these 2 studies did not alter the overall results.

In conclusion, the current study is the first meta-analysis to evaluate the effect of p53 status on predicting the response to chemotherapy-based in patients with esophageal cancer. Our meta-analysis suggested that p53 status, as a useful biomarker, was associated with the response to chemotherapy-based treatment, especially to NCRT in patients with ESCC. However, future large-scale and well-designed prospective studies are warranted to verify our finding and its value in clinical practice.

Acknowledgements

This work was supported by of Chinese Ministry of Health Key Program grant (No. 179). We would like to thank the authors of the studies included in our manuscript.

Supplementary material

10434_2012_2859_MOESM1_ESM.docx (29 kb)
Supplementary material 1 (DOCX 28 kb)
10434_2012_2859_MOESM2_ESM.tif (54 kb)
Supplement Figure 1: Forest plots of RR were estimated for association between p53 status and pathological complete response (CR) to neoadjuvant chemoradiotherapy (NCRT) in patients with esophageal squamous cell carcinoma. Supplementary material 2 (TIFF 53 kb)
10434_2012_2859_MOESM3_ESM.tif (67 kb)
Supplement Figure 2: The funnel plot shows that there was no obvious indication of publication bias for the outcome of total major response (MR) to chemotherapy-based treatment in patients with esophageal cancer. Supplementary material 3 (TIFF 66 kb)

Copyright information

© Society of Surgical Oncology 2013

Authors and Affiliations

  • Shui-Shen Zhang
    • 1
    • 2
    • 3
  • Qing-Yuan Huang
    • 1
    • 2
    • 3
  • Hong Yang
    • 1
    • 2
    • 3
  • Xuan Xie
    • 1
    • 2
    • 3
  • Kong-Jia Luo
    • 1
    • 2
    • 3
  • Jing Wen
    • 1
    • 2
    • 3
  • Xiao-Li Cai
    • 4
  • Fu Yang
    • 1
    • 2
    • 3
  • Yi Hu
    • 1
    • 2
    • 3
  • Jian-Hua Fu
    • 1
    • 2
    • 3
  1. 1.State Key Laboratory of Oncology in South ChinaSun Yat-Sen University Cancer CenterGuangzhouPeople’s Republic of China
  2. 2.Guangdong Esophageal Cancer Research InstituteGuangzhouPeople’s Republic of China
  3. 3.Department of Thoracic OncologySun Yat-Sen University Cancer CenterGuangzhouPeople’s Republic of China
  4. 4.Department of Oncology, Nanfang HospitalSouthern Medical UniversityGuangzhouPeople’s Republic of China

Personalised recommendations