- Email: [email protected]
Impact of STAT3 Phosphorylation on the Clinical Effectiveness of Anti-EGFR–Based Therapy in Patients With Metastatic Colorectal Cancer
Original Study
Impact of STAT3 Phosphorylation on the Clinical Effectiveness of Anti-EGFR–Based Therapy in Patients With Metastatic Colorectal Cancer Erion Dobi,1,2 Franck Monnien,3 Stefano Kim,1 Arben Ivanaj,4 Thiery N’Guyen,1 Martin Demarchi,1 Olivier Adotevi,1,2 Antoine Thierry-Vuillemin,1 Marie Jary,3 Bernadette Kantelip,3 Xavier Pivot,1,2 Yann Godet,2 Severine Valmary Degano,3 Christophe Borg1,2 Abstract Half of patients with KRAS wild-type colorectal cancer do not benefit from adding anti– epithelial growth factor receptor (EGFR) to standard chemotherapy regimens. This retrospective study was performed in 94 patients with metastatic colorectal cancer (mCRC) treated in the second line with cetuximab and chemotherapy. Signal transducer and activator of transcription 3 (STAT3) phosphorylation in tumor cells was correlated with decreased median progression-free survival and overall survival (OS). These results highlight the potential role of STAT3 as a molecular target to optimize anti-EGFR therapies. Background: Signal transducer and activator of transcription 3 (STAT3) is involved in epithelial growth factor receptor (EGFR) signaling in a KRAS-independent manner. Phosphorylated STAT3 (pSTAT3) expression in patients with metastatic colorectal cancer (mCRC) treated with anti-EGFR– containing salvage chemotherapy has never been investigated. Patients and Methods: The first endpoint of this retrospective study was to evaluate the impact of pSTAT3 on the time to progression (TTP) in 94 patients with mCRC treated with anti-EGFR– based therapies in the second- or third-line setting between July 2004 and November 2009. The influence of pSTAT3 on objective response rate and overall survival (OS) was also reported. Nuclear expression of pSTAT3 status was evaluated by immunohistochemical tests on formalin-fixed and paraffin-embedded tumor samples obtained before therapy. Results: Positive expression of pSTAT3 was observed in 24.5% of the tumor samples. The probability of achieving an objective response was 13% among patients with positive nuclear expression of pSTAT3 compared with 41% for patients displaying pSTAT3-negative tumors (P ⫽ .02). In a multivariate logistic regression model, high-grade skin rash, wild-type KRAS status, and negative pSTAT3 status significantly improved TTP and OS. Conclusion: These results underscore an impact of pSTAT3 on the clinical efficacy of anti-EGFR– containing chemotherapy regimens and support the prospective assessment of this biomarker. Clinical Colorectal Cancer, Vol. 12, No. 1, 28-36 © 2013 Elsevier Inc. All rights reserved. Keywords: Cetuximab, Colorectal cancer, EGFR, KRAS, Signal transducer and activator of transcription 3 (STAT3)
Introduction Despite the progress made in the management of metastatic colorectal cancer (mCRC) over the past few years, this disease remains an 1 Department of Medical Oncology, University Hospital of Besançon, Besançon, France 2 Institut National de la Sante et de la Recherche Medicale UMR 1098, University of Franche-Comté, Besançon, France 3 Department of Pathology, University Hospital of Besançon, Besançon, France 4 Department of Internal Medicine, University Hospital of Tirana, Tirana, Albania
Submitted: Mar 23, 2012; Revised: Aug 19, 2012; Accepted: Sep 13, 2012; Epub: Oct 17, 2012 Address for correspondence: Christophe Borg, MD, Medical Oncology Unit, J. Minjoz University Hospital, Besançon, F-25000, France Fax: ⫹33 3 81 66 88 58; e-mail contact: [email protected]
28
Clinical Colorectal Cancer March 2013
important health problem.1 Targeted therapies improved the clinical outcomes achieved by conventional chemotherapy regimens.2,4 In particular, targeting the extracellular domain of epithelial growth factor (EGFR) with monoclonal antibodies is an effective strategy for mCRC treatment either when used as a single agent5 or in combination with chemotherapy.3,4 The presence of KRAS mutations was identified as a predictive factor accounting for anti-EGFR resistance.6-9 Nevertheless, not all patients with KRAS wild-type (KRAS WT) tumors benefit from anti-EGFR treatments, suggesting that molecular biomarkers correlated to the clinical interest of anti-EGFR still need to be characterized.6-9 We therefore searched for molecular signaling involved both in mCRC oncogenesis and EGFR signaling and identified signal trans-
1533-0028/$ - see frontmatter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.clcc.2012.09.002
Figure 1 Immunohistochemical Analysis (ⴛ 200) of pSTAT3 Expression on Colorectal Cancers. (A) Positive Nuclear Staining in Tumor Cells. (B) Negative Nuclear Staining in Tumor Cells. Arrows Indicate Endothelial Cells, Arrowheads Indicate Lymphocytes, and Asterisks Indicate Tumor Cells
A
ducer and activator of transcription 3 (STAT3) as a potential predictive factor. STAT3 is a cytoplasmic transcription factor that might translocate into the nucleus after growth factor stimulation. In contrast to normal cells in which phosphorylation is transient, constitutive activation of STAT3 has been reported in several primary cancers and tumor cell lines.10-13 Phosphorylation allows nuclear translocation and DNA binding of the STAT3 dimer, leading to upregulation of several genes involved in the cell cycle and survival such as CCND1, MYC, or BCL2L1, as well as inhibition of ligands controlling innate immunity activation.13-17 Several reports have implicated STAT3 in colon cancer oncogenesis.10-12 In particular, STAT3 was formally identified as a molecular link between chronic intestinal inflammatory diseases and CRCs.18 It has been shown that STAT3 overexpression was associated with higher CRC-specific mortality, in line with a direct role of STAT3 in mCRC prognosis.19 Another function of STAT3 in mCRC might be the activation of molecular networks leading to chemotherapy resistance. Indeed, STAT3 signaling was shown to regulate EME1 endonuclease to reduce DNA damage.20 Besides EGFR,21 several other signaling pathways such as interleukin-618 or Src kinases12 might induce STAT3 activation in cancer cells and prevent the ability of anti-EGFR to counteract STAT3related oncogenic properties. We decided to assess the clinical outcomes of anti-EGFR– containing chemotherapy regimens in patients with mCRC according to pSTAT3 status.
Patients and Methods Patient Population Ninety-four consecutive patients with histologically confirmed mCRC and available tumor material for molecular analysis, who were treated with cetuximab-containing chemotherapy in secondline treatment or beyond, at the University Hospital of Besançon (Franche-Comté, France) between July 2004 and December 2009
B
were enrolled. Tumor location, World Health Organization (WHO) performance status, stage at diagnosis, number of metastases, higher grade of skin rash during treatment, sampling method of tumors, and chemotherapy administrated with cetuximab were included in the database. KRAS status was routinely ascertained from July 2008.
Immunohistochemical Analysis Formalin-fixed and paraffin-embedded tumor samples obtained before therapy were cut in 4-m sections using standard techniques. Immunostaining for pSTAT3 (Tyr705) was performed on the automated Benchmark XT automated slide preparation system (Ventana Medical Systems, Inc, Oro Valley, AZ) using the rabbit polyclonal antibody pSTAT3 (sc-7993, dilution 1:200; Santa Cruz Biotechnology Inc, Santa Cruz, CA). Briefly, sections were deparaffinized and antigen retrieval was carried out by incubating slides for 60 minutes in Cell Conditioning 1 (CC1) buffer. Sections were then incubated for 32 minutes with the primary rabbit polyclonal antibody pSTAT3 followed by biotin-labeled secondary anti–rabbit antibody for 8 minutes, and finally a solution of streptavidin-peroxidase for 8 minutes. Tissues were then stained for 8 minutes with diaminobenzidine substrate. Some serial sections were also incubated with antibody diluent only to provide negative controls. The evaluation of pSTAT3 expression was performed at high magnification (⫻400) on the tumor material obtained just before receiving cetuximab-containing treatment (biopsy samples or resection results of primary tumor or metastasis). Five carcinomatous fields were selected to count the number of cancer cells exhibiting nuclear staining. As previously described in colorectal carcinoma, biopsy samples were considered positive for pSTAT3 if nuclear staining was detected in more than 15% of tumor cells.22 Endothelial cells, which often stain positive for pSTAT3, served as internal positive controls. Lymph nodes from inflammatory diseases were used as external positive controls to set up pSTAT3 staining. Cytoplasmic staining was not taken into account. Figure 1 illustrates positive and
Clinical Colorectal Cancer March 2013
29
STAT3 Predicts Anti-EGFR-Based Therapy Efficiency Table 1 Patient Characteristics pSTAT3 Variable
No. (%) —
ⴙ
P Value
Sex Male
50 (53.2)
41 (57.7)
9 (39.1)
Female
44 (46.8)
30 (42.3)
14 (60.9)
.15 Median Age (y) (Range)
65.6 (29-91)
Age ⱕ 70
60 (63.8)
43 (60.6)
17 (73.9)
Age ⬎ 70
34 (36.2)
28 (39.4)
6 (26.1)
0
45 (47.9)
33 (46.5)
12 (52.2)
1
38 (40.4)
31 (43.7)
7 (30.4)
2
11 (11.7)
7 (09.9)
4 (17.4)
Resection of primary tumor
31 (33.0)
24 (33.8)
7 (30.4)
Resection of metastasis
31 (33.0)
26 (36.6)
5 (21.7)
Biopsy of primary tumor
15 (16.0)
11 (15.5)
4 (17.4)
Biopsy of metastasis
17 (18.0)
10 (14.1)
7 (30.4)
Colon
59 (62.8)
48 (67.6)
11 (47.8)
Rectum
35 (37.2)
23 (32.4)
12 (52.2)
.32
ECOG Performance Status
.42
Sampling Method
.28
Tumor Location .13 Stage at Diagnosis III
19 (20.2)
15 (21.1)
4 (17.4)
III
26 (27.7)
20 (28.2)
6 (26.1)
IV
49 (52.1)
36 (50.7)
13 (56.5)
1
29 (30.9)
24 (33.8)
5 (21.7)
ⱖ2
65 (69.1)
47 (66.2)
18 (78.3)
Grade 0/1
34 (37.8)
26 (38.2)
8 (36.4)
Grade 2/3
56 (62.2)
42 (61.8)
14 (63.6)
Before 09/2005
48 (51.0)
39 (54.9)
9 (39.2)
After 10/2005
46 (49.0)
32 (45.1)
14 (60.8)
Negative
29 (30.9)
29 (40.9)
0 (0.0)
Positive
65 (69.1)
42 (59.1)
23 (100.0)
Wild type
61 (74.4)
48 (77.4)
13 (65.0)
Mutant
21 (25.6)
14 (22.6)
7 (35.0)
Second
57 (60.6)
42 (59.2)
15 (65.2)
Third
37 (39.4)
29 (40.8)
8 (34.8)
No
24 (25.5)
15 (21.1)
9 (39.1)
Yes
70 (74.5)
56 (78.9)
14 (60.9)
Irinotecan-based
84 (89.4)
62 (87.3)
22 (95.7)
Oxaliplatin-based
10 (10.6)
9 (12.7)
1 (4.3)
.87
Number of Metastases .31
Skin Toxicity .99 Date of Fixation .3 pSTAT3 Status in TILs < .01 KRAS Status .37 Cetuximab Administration Line .63 Exposure to Irinotecan Before Cetuximab .1 Chemotherapy Regimen Administrated With Cetuximab .44
Abbreviations: ECOG ⫽ Eastern Cooperative Oncology Group; pSTAT3 ⫽ phosphorylated signal transducer and activator of transcription 3; TILs ⫽ tumor infiltrating lymphocytes.
30
Clinical Colorectal Cancer March 2013
Erion Dobi et al Table 2 Evaluation of Treatment Efficacy Regarding RECIST Criteria Variable
No. (%)
Response a
OR
SD
PD
11 (32.4)
4 (14.3)
P Value
Stage at Diagnosis II
19 (20.2)
4 (12.5)
III
26 (27.7)
13 (40.6)
6 (17.6)
7 (25.0)
IV
49 (52.1)
15 (46.9)
17 (50.0)
17 (60.7)
1
29 (30.9)
10 (31.3)
13 (38.2)
6 (21.4)
ⱖ2
65 (69.1)
22 (68.7)
21 (61.8)
22 (78.6)
.10
Number of Metastases .36
Cutaneous Toxicity Grade 0/1
34 (37.8)
7 (23.3)
10 (30.3)
17 (63.0)
Grade 2/3
56 (62.2)
23 (76.7)
23 (69.7)
10 (37.0)
Negative
71 (75.5)
29 (90.6)
21 (61.8)
21 (75.0)
Positive
23 (24.5)
3 (9.4)
13 (38.2)
7 (25.0)
Negative
29 (30.9)
14 (40.7)
8 (23.5)
7 (25.9)
Positive
65 (69.1)
19 (59.3)
26 (76.5)
20 (74.1)
Wild type
61 (74.4)
27 (93.1)
19 (65.5)
15 (62.5)
Mutant
21 (25.6)
2 (6.9)
10 (34.5)
9 (37.5)
2
57 (60.6)
19 (59.4)
20 (58.8)
18 (64.3)
3
37 (39.4)
13 (40.6)
14 (41.2)
10 (35.7)
No
24 (25.5)
8 (25.0)
8 (23.5)
8 (28.6)
Yes
70 (74.5)
24 (75.0)
26 (76.5)
20 (71.4)
< .01
pSTAT3 Status in Tumor Cells .02
pSTAT3 Status in TILs .13
KRAS Status .02
Cetuximab Administration Line .89
Exposure to Irinotecan Before Cetuximab .89
Abbreviations: CR ⫽ complete response; OR ⫽ objective response; PD ⫽ progressive disease; PR ⫽ partial response; pSTAT3 ⫽ phosphorylated signal transducer and activator of transcription 3; SD ⫽ stable disease; TILs ⫽ tumor-infiltrating lymphocytes. a OR ⫽ PR ⫹ CR.
negative pSTAT3 staining. Of note, pSTAT3 expression was not affected by the length of storage (Table 1). Analysis of the immunohistochemical staining was interpreted by 2 independent pathologists. In case of discrepancies, the pathologists reviewed the slides together and reached consensus.
Multiplex PCR products were checked for quality and yield by running 18 L in 2% agarose-Tris-borate electrophoresis gel. After purification using the gel extraction kit NucleoSpin Extract II (Macherey-Nagel, Hoerdt, France), PCR products were analyzed for the presence of KRAS codon 12 and codon 13 mutations at nucleotides c.34, c.35, c.37, and c.38.
Samples and DNA Extraction and KRAS Mutation Analysis
Statistical Analysis
Formalin-fixed and paraffin-embedded tissues containing at least 60% tumor cells were selected, and DNA extraction was performed by using the QIAamp DNA Mini Kit (Qiagen, Courtaboeuf, France). Briefly, polymerase chain reaction (PCR) was performed in a 25-L reaction containing 1⫻ Qiagen Multiplex PCR Master Mix (Qiagen, Courtaboeuf, France), Plzf and Af4 primers at 0.5-M final concentration, and Tbxas and Rag1 primers at a final concentration of 0.25 M. A multiplex PCR assay was designed to amplify fragments of KRAS exon 2 (200 bp) and to identify activating mutations using a SNaPshot assay.23
Quantitative variables were recoded into categorical variables. All categorical data were compared using the Fisher exact test or 2 test. Time to progression (TTP) and overall survival (OS) were measured from the date of initiation of the first cetuximab-containing treatment line to the first recurrence or death, respectively. The endpoint date was December 31, 2010. TTP and OS were calculated with the Kaplan-Meier method. A search for parameters related to TTP and OS length was performed by univariate analysis using the log-rank test. The univariate analysis considered the following variables: sex, age (ⱕ 70 vs.
Clinical Colorectal Cancer March 2013
31
STAT3 Predicts Anti-EGFR-Based Therapy Efficiency Table 3 TTP and OS in Second-Line or Beyond Cetuximab-Containing Treatment According to Clinicopathologic Characteristics Variable
No. (%)
TTP (Mo) Median (95% CI)
P Value
OS (Mo) Median (95% CI)
P Value
Sex Male
50 (53.2)
5.3 (4.3-6.3)
Female
44 (46.8)
6.2 (5.1-7.3)
Median Age (y) (Range)
.37
11.9 (8.0-15.8) 12.0 (9.5-14.6)
.31
65.6 (29-91)
Age ⱕ 70
60 (63.8)
5.7 (4.5-6.8)
Age ⬎ 70
34 (36.2)
5.7 (4.4-6.9)
0
45 (47.9)
6.2 (4.6-7.8)
1
38 (40.4)
5.7 (2.6-8.6)
2
11 (11.7)
2.9 (1.1-6.1)
Colon
59 (62.8)
6.7 (5.2-8.1)
Rectum
35 (37.2)
5.2 (4.0-6.3)
.85
12.4 (10.7-14.1) 10.0 (5.2-14.8)
.56
ECOG Performance Status 15.0 (11.1-19.2) .08
10.7 (6.9-14.5)
.01
5.8 (1.8-9.7)
Tumor Location .12
12.1 (9.9-14.4) 11.0 (6.7-15.3)
.53
Stage at Diagnosis II
19 (20.2)
5.7 (3.2-8.1)
III
26 (27.7)
6.1 (3.0-9.2)
IV
49 (52.1)
5.6 (4.6-6.7)
1
29 (30.9)
6.2 (5.1-7.3)
ⱖ2
65 (69.1)
5.6 (4.6-6.5)
Grade 1/2
34 (37.8)
3.3 (1.0-5.6)
Grade 2/3
56 (62.2)
6.3 (5.2-7.3)
14.9 (6.4-23.4) .34
14.0 (9.4-18.8)
.15
10.0 (7.3-12.7)
Number of Metastases .50
14.5 (9.1-19.6) 11.8 (8.8-14.8)
.38
Cutaneous Toxicity < .01
5.1 (4.0-6.2) 14.8 (11.2-18.5)
< .01
pSTAT3 Status in Tumor Cells Negative
71 (75.5)
6.3 (5.1-7.6)
Positive
23 (24.5)
4.5 (4.1-4.9)
Negative
29 (31.5)
6.2 (3.4-9.0)
Positive
63 (68.5)
5.7 (4.1-7.2)
< .01
13.1 (7.6-18.7) 9.4 (7.2-11.6)
.02
pSTAT3 Status in TILs .05
14.5 (6.6-22.4) 12.0 (9.8-14.3)
.37
KRAS Status Wild type
61 (74.4)
6.6 (5.3-7.8)
Mutant
21 (25.6)
3.9 (2.6-5.2)
Second
57 (60.6)
5.6 (4.6-6.6)
Third
37 (39.4)
6.0 (4.9-7.1)
< .0001
13.5 (10.0-16.9) 8.8 (4.9-12.4)
< .01
Cetuximab Administration Line .81
12.0 (8.9-15.1) 13.0 (8.0-17.9)
.86
Exposure to Irinotecan Before Cetuximab No
24 (25.5)
5.1 (3.7-6.5)
Yes
70 (74.5)
6.1 (5.1-7.1)
.23
12.7 (4.9-20.5) 11.8 (9.2-14.4)
.51
Chemotherapy Regimen Administered With Cetuximab Irinotecan-based
84 (89.4)
5.6 (4.8-6.3)
Oxaliplatin-based
10 (10.6)
7.2 (5.5-8.9)
.95
11.0 (8.4-13.7) 21.9 (0.9-42.9)
.34
Abbreviations: CI ⫽ confidence interval; ECOG ⫽ Eastern Cooperative Oncology Group; OS ⫽ overall survival; pSTAT3 ⫽ phosphorylated signal transducer and activator of transcription 3; TILs ⫽ tumor infiltrating lymphocytes; TTP ⫽ time to progression.
32
Clinical Colorectal Cancer March 2013
Erion Dobi et al Table 4 Multivariate Analysis for Objective Response Variable
Response Rate, No. (%)
Odds Ratio (95% CI)
P Value
KRAS (Wild-Type)
27/61 (44.3%)
0.17 (0.04-0.83)
.03
PSTAT3 (Negative)
29/71 (40.8%)
0.19 (0.04-0.91)
.04
Abbreviation: pSTAT3 ⫽ phosphorylated signal transducer and activator of transcription 3.
⬎ 70 years), WHO performance status (0 vs. 1-2), tumor location (colon vs. rectum), stage at diagnosis (stage II vs. III vs. stage IV), number of metastases before cetuximab initiation (1 vs. ⱖ 2), skin rash (grades 0/1 vs. grades 2/3), nuclear pSTAT3 status in tumor cells (negative vs. positive), pSTAT3 status in tumor-infiltrating lymphocytes (negative vs. positive), KRAS status, cetuximab line of treatment (second vs. third), chemotherapy administrated with cetuximab (irinotecan-based vs. oxaliplatin-based chemotherapy) and if the patients were exposed to irinotecan before cetuximabcontaining treatment. Odds ratios were estimated for objective response by logistic regression model. Multivariate analysis was performed to identify independent prognostic factors for survival using a Cox proportional hazard model with forward stepwise selection. The significance levels for entry and for stay were 0.2 and 0.1, respectively. Statistical significance for all analyses in the present report is considered at the 2-sided 5% level. The data analysis was generated using SPSS software, version 18.0.0 (IBM Corp, Armonk, NY).
Results Patients Table 1 shows the characteristics of the 94 patients. All patients were treated with cetuximab in combination with chemotherapy (89.4% with irinotecan, 10.6% with oxaliplatin) as second-line treatment or beyond. There were no patients who received the antiEGFR treatment in the first-line setting. Fifty-seven (60.6%) patients received the treatment in the second-line setting, and the remaining 37 (39.4%) patients had it as third-line treatment. Seventy (74.5%) patients had been exposed to irinotecan before cetuximabcontaining treatment. The immunohistochemical expression of nuclear pSTAT3 in tumor cells was determined in all 94 patients with mCRC, whereas KRAS status was determined in the 82 patients for whom tumor material was available. The nuclear pSTAT3 expression and KRAS mutations were detected in 23 (24.5%) and 21 (25.6%) of the tumor samples, respectively. The median follow-up was 22 months and all patients were assessable for clinical objective response, TTP, and OS. Disease characteristics were compatible with common features of patients with mCRC. Clinical outcomes in this cohort of patients were also in agreement with previously reported results. Regarding cetuximab therapy, grade ⱖ 2 acneiform rash occurred in 62% of the patients. Overall response rate assessed by the Response Evaluation Criteria in Solid Tumors (RECIST) criteria was observed in 33% of the patients (Table 2). TTP and OS were 5.8 and 12.5 months, respectively (Table 3). Of note, classic predictive factors were significantly associated with objective response rate (ORR) and TTP in this cohort. Indeed, ORR was increased for patients with KRAS WT
tumors (44% vs. 9% in the case of KRAS mutant tumors; P ⫽ .02) and TTP was 6.6 months for these patients compared with 3.9 months for patients whose tumors were KRAS mutant. As seen in Tables 2 and 3, skin rash was also associated with ORR (41% vs. 21%) and TTP (6.3 vs. 3.3 months).
Immunohistological Analysis of pSTAT3 Status The kappa value for concordance between both pathologists for evaluating pSTAT3 status was 0.86. To control for potential bias, the impact of the sampling method on the nuclear immunohistochemical expression of pSTAT3 was monitored. There was no significant difference dependent on the tumor sampling method (P ⫽ .36). With a cutoff of 15%, we identified 23 patients (24.5%) whose tumors expressed high levels of pSTAT3 within the nucleus of tumor cells. As shown in Table 1, we did not observe any significant influence between the patients characteristics and the expression of pSTAT3. Moreover, the status of pSTAT3 did not differ in KRAS-WT or KRAS-mutant mCRC (Table 1). Interestingly, tumorinfiltrating lymphocytes (TILs) expressing pSTAT3 were observed in all 23 pSTAT3-positive tumors, whereas only 59% of pSTAT3negative tumors were associated with pSTAT3-positive TIL.
Influence of pSTAT3 Status on Tumor Response to Cetuximab-containing Chemotherapy ORR was significantly lower among patients with positive nuclear expression of pSTAT3 regarding RECIST criteria. The probability of achieving an objective response was 13% when a positive nuclear expression of pSTAT3 was reported, compared with 41% for patients displaying pSTAT3-negative tumors (P ⫽ .02) (Table 2). In the multivariate logistic regression model, pSTAT3-negative status and KRAS WT status both remained significant to predict overall responses occurring in the context of cetuximab-containing chemotherapy regimens (odds ratio [OR], 0.17; 95% confidence interval [CI], 0.04-0.83; P ⫽ .04; OR, 0.19; 95% CI, 0.04-0.91; P ⫽ .03, respectively) (Table 4).
Influence of pSTAT3 Status on Time to Progression In univariate analysis regarding TTP, patients with expression of pSTAT3 in tumor cells had a worse prognosis (4.5 vs. 6.3 months; P ⬍ .01) (Table 3; Figure 2A). Among patients with WT KRAS tumors, expression of nuclear pSTAT3 in tumor cells remained significantly associated with worse TTP (5.1 vs. 7.1 months, P ⫽ .04). pSTAT3 did not influence TTP in patients whose tumors displayed KRAS mutations (4.5 vs. 3.3 months, P ⫽ .52; Table 5). In multivariate analysis, the following factors remained significant for length of TTP: KRAS WT status (hazard ratio [HR], 0.4; 95% CI, 0.2-0.7; P ⬍ .01), negative pSTAT3 status (HR, 0.5; 95% CI, 0.3-.09; P ⫽ .02), and high-grade skin rash (HR, 0.6; 95% CI, 0.4-.09; P ⫽ .02) (Table 6).
Influence of pSTAT3 Status on Overall Survival In univariate analysis regarding OS, patients with positive nuclear expression of pSTAT3 in tumor cells had a worse OS (9.4 vs. 13.1 months; P ⫽ .02) (Figure 2B). OS was significantly lower among patients whose tumors carried KRAS mutations (8.8 vs. 13.5 months; P ⬍ .01).
Clinical Colorectal Cancer March 2013
33
STAT3 Predicts Anti-EGFR-Based Therapy Efficiency Figure 2 Kaplan-Meier Curves According to pSTAT3 Status for Time to Tumor Progression (A) and Overall Survival (B)
A 1.0
B 1.0 pSTAT3 status negative positive
0.8
0.6
Probability
Probability
0.8
pSTAT3 status negative positive
0.4
0.2
0.6
0.4
0.2 Log-rank P < 0.01
Log-rank P = 0.02
0.0
0.0 0
No. at risk Negative 71 Positive 23
6 12 18 Time to tumor progression (months)
44 8
12 0
3 0
24
0
1 0
No. at risk Negative 71 Positive 23
6 12 18 Overall survival (months)
24
52 16
18 2
37 10
25 2
Table 5 TTP and OS in Second-Line or Beyond Cetuximab-Containing Treatment According to pSTAT3 Status for KRAS WT and KRAS MT Variable
No.
TTP (mo) Median (95% CI)
P Value
OS (mo) Median (95% CI)
P Value
pSTAT3 Status in KRAS WT Tumor Cells Negative
48
7.1 (5.6-8.5)
Positive
13
5.1 (3.7-6.5)
Negative
14
3.3 (1.5-5.1)
Positive
7
4.5 (4.2-4.9)
.04
16.6 (9.8-23.4) 12.7 (7.3-13.2)
.13
pSTAT3 Status in KRAS MT Tumor Cells .52
7.3 (0.3-14.4) 8.8 (0.2-17.4)
.34
Abbreviations: MT ⫽ mutant; OS ⫽ overall survival; pSTAT3 ⫽ phosphorylated signal transducer and activator of transcription 3; TTP ⫽ time to progression; WT ⫽ wild type.
In multivariate analysis regarding OS, the independent significant prognostic factors were high-grade skin rash (HR, 0.5; 95% CI, 0.3-0.8; P ⬍ .01), KRAS WT status (HR, 0.5; 95% CI, 0.3-0.9; P ⫽ .02), and negative pSTAT3 status (HR, 0.5; 95% CI, 0.3-0.9; P ⫽ .03).
Discussion The addition of anti-EGFR monoclonal antibodies to conventional therapies extended the therapeutic strategies available in mCRC management. EGFR targeting in this disease has been demonstrated as a relevant option to improve overall response along with to standard chemotherapy regimens3,24 or to restore the efficacy of irinotecan in previously treated patients with mCRC.25 Although the presence of KRAS mutations is a specific predictive biomarker for lack of anti-EGFR monoclonal antibody efficacy, there is convincing evidence that additional events are involved in this process, since not all patients with KRAS WT tumors benefit from anti-EGFR mono-
34
Clinical Colorectal Cancer March 2013
clonal antibodies.25,26 We have demonstrated, for the first time, that activated STAT3 (pSTAT3) may contribute, along with KRAS status, to the selection of patients eligible for anti-EGFR– based therapies. Patient characteristics and clinical outcomes observed in our cohort are consistent with previously published literature and with patients treated in general practice. KRAS molecular status and skin rash were correlated to cetuximab efficacy. In the present study, 24.5% of the tumor samples displayed a high nuclear expression of pSTAT3. This observation is in agreement with the 18% level of pSTAT3 expression reported recently in mCRC.19 The negative impact of pSTAT3 on clinical outcomes achieved by anti-EGFR therapy is consistent with the pivotal role of STAT3 signaling in oncogenesis.27 Several mechanisms driven by the transcriptional activity of STAT3 might promote cancer progression.
Erion Dobi et al Table 6 Multivariate Analysis for TTP and OS Variable
HR (95% CI)
P Value
KRAS (wild type)
0.4 (0.2-0.7)
< .01
Skin rash (grade 2/3)
0.6 (0.4-0.9)
.02
pSTAT3 (negative)
0.5 (0.3-0.9)
.02
Skin rash (grade 2/3)
0.5 (0.3-0.8)
< .01
KRAS (wild type)
0.5 (0.3-0.9)
.02
pSTAT3 (negative)
0.5 (0.3-0.9)
.03
●
TTP ● ●
OS
Abbreviations: OS ⫽ overall survival; pSTAT3 ⫽ phosphorylated signal transducer and activator of transcription 3; TTP ⫽ time to progression.
Indeed, STAT3 directly induces cyclin D1 and vascular endothelial growth factor transcription in tumor cells, leading to cancer proliferation and angiogenesis.15,28 The adverse prognostic value of STAT3 was recently confirmed in patients with mCRC. Morikawa et al established a negative correlation between pSTAT3 expression in CRC cells and patient survival.19 A possible hypothesis accounting for the predictive value of STAT3 might be the ability of this signaling pathway to confer resistance to chemotherapy. In vitro, STAT3 activation was shown to be related to chemotherapy resistance in B-cell lymphoma,29 multiple myeloma,30 and CRC.21 STAT3 transcriptional activity increases BCL2L1 and MCL1 expression, leading to drug resistance by inhibiting the cell death pathways associated with genotoxic treatment. Other experiments involved the interleukin-6/Jak2/STAT3 pathway in cancer stem cell homeostasis, which had been shown to include most of the chemotherapy-resistant cells.31-33 In the clinical setting, the presence of pSTAT3 in breast cancer cells is correlated to a lower probability of complete pathologic response after neoadjuvant chemotherapy, sustaining a possible role for pSTAT3 in chemotherapy resistance.34 The potential involvement of STAT3 signaling in chemotherapy resistance is illustrated in our study by a decreased ORR achieved by anti-EGFR and chemotherapy when pSTAT3 status was observed in cancer cells (Table 2). These exploratory results prompt the development of further clinical studies to confirm the predictive value of pSTAT3 on anti-EGFR therapy effectiveness. It would be of particular interest to confirm the clinical relevance of these results by comparing the impact of pSTAT3 in a controlled randomized trial assessing the efficacy of chemotherapy with or without anti-EGFR. Furthermore, the role of STAT3 as a molecular target to optimize anti-EGFR therapies was recently highlighted.35 A preclinical study demonstrated that in contrast to cetuximab monotherapy, dual inhibition of EGFR signaling using cetuximab and erlotinib inhibits pSTAT3.35 Inhibition of STAT3 in mCRCs might be a promising strategy to target chemotherapy-resistant cancer cells and should be assessed in association with anti-EGFR monoclonal antibodies.
Clinical Practice Points ●
Targeting the extracellular domain of EGFR with monoclonal antibodies (cetuximab or panitumumab) is an effective strategy as
●
● ●
a single agent or in combination with chemotherapy regimens for patients with mCRC and KRAS WT tumors. Currently, KRAS codon 12 and codon 13 mutations are the only predictive biomarkers used in clinical practice to prescribe antiEGFR antibodies. In this retrospective study, we observed an activation of STAT3 in 24.5% of the tumor samples analyzed. Our results showed a negative impact of pSTAT3 in the ORR achieved by an anti-EGFR and chemotherapy regimen combination and decreased progression-free and OS in patients with mCRC displaying an active form of STAT3 in their tumors. Immunohistochemical analysis of pSTAT3 status might be of clinical interest to better individualize the potential benefit of antiEGFR therapies. The predictive value of this biomarker should be confirmed in further randomized clinical trials. Moreover, these results sustain the development of preclinical and clinical research to assess the impact of STAT3 molecular inhibition to potentiate anti-EGFR therapies.
Acknowledgments We are grateful to Ms Alice Baraquin, Ms Marthe Bigand, Ms Adeline Bouard, Mr Christophe Bracieux, Mr Benoît Girardo, and Ms Monique Guyot from the Department of Molecular Biology and Pathology for their excellent technical assistance with tumor section preparation and immunohistochemical and biological analysis. We also thank the biobank Tumorothèque Régionale de Franche-Comté for making available all useful tumor samples for the research. This work was supported by a research grant from the “Ligue contre le cancer du Grand Est”.
Disclosure The authors have stated that they have no conflicts of interest.
References 1. Hurwitz H, Fehrenbacher L, Novotny W, et al. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 2004; 350:2335-42. 2. Van Cutsem E, Köhne CH, Láng I, et al. Cetuximab plus irinotecan, fluorouracil, and leucovorin as first-line treatment for metastatic colorectal cancer: updated analysis of overall survival according to tumor KRAS and BRAF mutation status. J Clin Oncol 2011; 29:2011-9. 3. Douillard JY, Siena S, Cassidy J, et al. Randomized, phase III trial of panitumumab with infusional fluorouracil, leucovorin, and oxaliplatin (FOLFOX4) versus FOLFOX4 alone as first-line treatment in patients with previously untreated metastatic colorectal cancer: the PRIME study. J Clin Oncol 2010; 28:4697-705. 4. Van Cutsem E, Peeters M, Siena S, et al. Open-label phase III trial of panitumumab plus best supportive care compared with best supportive care alone in patients with chemotherapy-refractory metastatic colorectal cancer. J Clin Oncol 2007; 25:1658-64. 5. Lièvre A, Bachet JB, Le Corre D, et al. KRAS mutation status is predictive of response to cetuximab therapy in colorectal cancer. Cancer Res 2006; 66:3992-5. 6. Di Fiore F, Blanchard F, Charbonnier F, et al. Clinical relevance of KRAS mutation detection in metastatic colorectal cancer treated by cetuximab plus chemotherapy. Br J Cancer 2007; 96:1166-9. 7. Karapetis CS, Khambata-Ford S, Jonker DJ, et al. K-ras mutations and benefit from cetuximab in advanced colorectal cancer. N Engl J Med 2008; 359:1757-65. 8. Amado RG, Wolf M, Peeters M, et al. Wild-type KRAS is required for panitumumab efficacy in patients with metastatic colorectal cancer. J Clin Oncol 2008; 26:1626-34. 9. Yu H, Jove H. The STATs of cancer—new molecular targets come of age. Nat Rev Cancer 2004; 4:97-105. 10. Levy DE, Lee CK. What does Stat3 do? J Clin Invest 2002; 109:1143-8. 11. Bromberg JF, Horvath CM, Besser D, et al. Stat3 activation is required for cellular transformation by v-src. Mol Cell Biol 1998; 18:2553-8.
Clinical Colorectal Cancer March 2013
35
STAT3 Predicts Anti-EGFR-Based Therapy Efficiency 12. Bedel R, Thiery-Vuillemin A, Grandclement C, et al. Novel role for STAT3 in transcriptional regulation of NK immune cell targeting receptor MICA on cancer cells. Cancer Res 2011; 71:1615-26. 13. Barré B, Vigneron A, Coqueret O. The STAT3 transcription factor is a target for the Myc and riboblastoma proteins on the Cdc25A promoter. J Biol Chem 2005; 280: 15673-81. 14. Leslie K, Lang C, Devgan G, et al. Cyclin D1 is transcriptionally regulated by and required for transformation by activated signal transducer and activator of transcription 3. Cancer Res 2006; 66:2544-52. 15. Shirogane T, Fukada T, Muller JM, et al. Synergistic roles for Pim-1 and c-Myc in STAT3-mediated cell cycle progression and antiapoptosis. Immunity 1999; 11:709-19. 16. Kusaba T, Nakayama T, Yamazumi K, et al. Activation of STAT3 is a marker of poor prognosis in human colorectal cancer. Oncol Rep 2006; 15:1445-51. 17. Bollrath J, Phesse TJ, von Burstin VA, et al. gp130-mediated Stat3 activation in enterocytes regulates cell survival and cell-cycle progression during colitis-associated tumorigenesis. Cancer Cell 2009; 15:91-102. 18. Morikawa T, Baba Y, Yamauchi M, et al. STAT3 expression, molecular features, inflammation patterns, and prognosis in a database of 724 colorectal cancers. Clin Cancer Res 2011; 17:1452-62. 19. Vigneron A, Gamelin E, Coqueret O. The EGFR-STAT3 oncogenic pathway upregulates the Eme1 endonuclease to reduce DNA damage after topoisomerase I inhibition. Cancer Res 2008; 68:815-25. 20. Grandis JR, Drenning SD, Chakraborty A, et al. Requirement of Stat3 but not Stat1 activation for epidermal growth factor receptor–mediated cell growth in vitro. J Clin Invest 1998; 102:1385-92. 21. Barré B, Vigneron A, Perkins N, et al. The STAT3 oncogene as a predictive marker of drug resistance. Trends Mol Med 2007; 13:4-11. 22. Magnin S, Viel E, Baraquin A, et al. A multiplex SNaPshot assay as a rapid method for detecting KRAS and BRAF mutations in advanced colorectal cancers. J Mol Diagn 2011; 13:485-92. 23. Bokemeyer C, Bondarenko I, Hartmann JT, et al. Efficacy according to biomarker status of cetuximab plus FOLFOX-4 as first-line treatment for metastatic colorectal cancer: the OPUS study. Ann Oncol 2011; 22:1535-46.
36
Clinical Colorectal Cancer March 2013
24. Cunningham D, Humblet Y, Siena S, et al. Cetuximab monotherapy and cetuximab plus irinotecan in irinotecan-refractory metastatic colorectal cancer. N Engl J Med 2004; 351:337-45. 25. Wong R, Cunningham D. Using predictive biomarkers to select patients with advanced colorectal cancer for treatment with epidermal growth factor receptor antibodies. J Clin Oncol 2008; 26:5668-70. 26. Bromberg JF, Wrzeszczynska MH, Devgan G, et al. Stat3 as an oncogene. Cell 1999; 98:295-303. 27. Niu G, Wright KL, Huang M, et al. Constitutive Stat3 activity up-regulates VEGF expression and tumor angiogenesis. Oncogene 2002; 21:2000-8. 28. Alas S, Bonavida B. Rituximab inactivates signal transducer and activation of transcription 3 (STAT3) activity in B-non-Hodgkin’s lymphoma through inhibition of the interleukin 10 autocrine/paracrine loop and results in down-regulation of Bcl-2 and sensitization to cytotoxic drugs. Cancer Res 2001; 61:5137-44. 29. Alas S, Bonavida B. Inhibition of constitutive STAT3 activity sensitizes resistant non-Hodgkin’s lymphoma and multiple myeloma to chemotherapeutic drug-mediated apoptosis. Clin Cancer Res 2003; 9:316-26. 30. Marotta LL, Almendro V, Marusyk A, et al. The JAK2/STAT3 signaling pathway is required for growth of CD44⫹CD24⫺ stem cell-like breast cancer cells in human tumors. J Clin Invest 2011; 121:2723-35. 31. Creighton CJ, Li X, Landis M, et al. Residual breast cancers after conventional therapy display mesenchymal as well as tumor-initiating features. Proc Natl Acad Sci U S A 2009; 106:13820-5. 32. Li X, Lewis MT, Huang J, et al. Intrinsic resistance of tumorigenic breast cancer cells to chemotherapy. J Natl Cancer Inst 2008; 100:672-9. 33. Diaz N, Minton S, Cox C, et al. Activation of STAT3 in primary tumors from high-risk breast cancer patients is associated with elevated levels of activated SRC and survivin expression. Clin Cancer Res 2006; 12:20-8. 34. Ruff-Jamison S, Zhong Z, Wen Z, et al. Epidermal growth factor and lipopolysaccharide activate Stat3 transcription factor in mouse liver. J Biol Chem 1994; 269: 21933-45. 35. Weickhardt AJ, Price TJ, Chong G, et al. Dual targeting of the epidermal growth factor receptor using the combination of cetuximab and erlotinib: preclinical evaluation and results of the phase II DUX study in chemotherapy-refractory, advanced colorectal cancer. J Clin Oncol 2012; 30:1505-12.
Impact of STAT3 Phosphorylation on the Clinical Effectiveness of Anti-EGFR–Based Therapy in Patients With Metastatic Colorectal Cancer Erion Dobi,1,2 Franck Monnien,3 Stefano Kim,1 Arben Ivanaj,4 Thiery N’Guyen,1 Martin Demarchi,1 Olivier Adotevi,1,2 Antoine Thierry-Vuillemin,1 Marie Jary,3 Bernadette Kantelip,3 Xavier Pivot,1,2 Yann Godet,2 Severine Valmary Degano,3 Christophe Borg1,2 Abstract Half of patients with KRAS wild-type colorectal cancer do not benefit from adding anti– epithelial growth factor receptor (EGFR) to standard chemotherapy regimens. This retrospective study was performed in 94 patients with metastatic colorectal cancer (mCRC) treated in the second line with cetuximab and chemotherapy. Signal transducer and activator of transcription 3 (STAT3) phosphorylation in tumor cells was correlated with decreased median progression-free survival and overall survival (OS). These results highlight the potential role of STAT3 as a molecular target to optimize anti-EGFR therapies. Background: Signal transducer and activator of transcription 3 (STAT3) is involved in epithelial growth factor receptor (EGFR) signaling in a KRAS-independent manner. Phosphorylated STAT3 (pSTAT3) expression in patients with metastatic colorectal cancer (mCRC) treated with anti-EGFR– containing salvage chemotherapy has never been investigated. Patients and Methods: The first endpoint of this retrospective study was to evaluate the impact of pSTAT3 on the time to progression (TTP) in 94 patients with mCRC treated with anti-EGFR– based therapies in the second- or third-line setting between July 2004 and November 2009. The influence of pSTAT3 on objective response rate and overall survival (OS) was also reported. Nuclear expression of pSTAT3 status was evaluated by immunohistochemical tests on formalin-fixed and paraffin-embedded tumor samples obtained before therapy. Results: Positive expression of pSTAT3 was observed in 24.5% of the tumor samples. The probability of achieving an objective response was 13% among patients with positive nuclear expression of pSTAT3 compared with 41% for patients displaying pSTAT3-negative tumors (P ⫽ .02). In a multivariate logistic regression model, high-grade skin rash, wild-type KRAS status, and negative pSTAT3 status significantly improved TTP and OS. Conclusion: These results underscore an impact of pSTAT3 on the clinical efficacy of anti-EGFR– containing chemotherapy regimens and support the prospective assessment of this biomarker. Clinical Colorectal Cancer, Vol. 12, No. 1, 28-36 © 2013 Elsevier Inc. All rights reserved. Keywords: Cetuximab, Colorectal cancer, EGFR, KRAS, Signal transducer and activator of transcription 3 (STAT3)
Introduction Despite the progress made in the management of metastatic colorectal cancer (mCRC) over the past few years, this disease remains an 1 Department of Medical Oncology, University Hospital of Besançon, Besançon, France 2 Institut National de la Sante et de la Recherche Medicale UMR 1098, University of Franche-Comté, Besançon, France 3 Department of Pathology, University Hospital of Besançon, Besançon, France 4 Department of Internal Medicine, University Hospital of Tirana, Tirana, Albania
Submitted: Mar 23, 2012; Revised: Aug 19, 2012; Accepted: Sep 13, 2012; Epub: Oct 17, 2012 Address for correspondence: Christophe Borg, MD, Medical Oncology Unit, J. Minjoz University Hospital, Besançon, F-25000, France Fax: ⫹33 3 81 66 88 58; e-mail contact: [email protected]
28
Clinical Colorectal Cancer March 2013
important health problem.1 Targeted therapies improved the clinical outcomes achieved by conventional chemotherapy regimens.2,4 In particular, targeting the extracellular domain of epithelial growth factor (EGFR) with monoclonal antibodies is an effective strategy for mCRC treatment either when used as a single agent5 or in combination with chemotherapy.3,4 The presence of KRAS mutations was identified as a predictive factor accounting for anti-EGFR resistance.6-9 Nevertheless, not all patients with KRAS wild-type (KRAS WT) tumors benefit from anti-EGFR treatments, suggesting that molecular biomarkers correlated to the clinical interest of anti-EGFR still need to be characterized.6-9 We therefore searched for molecular signaling involved both in mCRC oncogenesis and EGFR signaling and identified signal trans-
1533-0028/$ - see frontmatter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.clcc.2012.09.002
Figure 1 Immunohistochemical Analysis (ⴛ 200) of pSTAT3 Expression on Colorectal Cancers. (A) Positive Nuclear Staining in Tumor Cells. (B) Negative Nuclear Staining in Tumor Cells. Arrows Indicate Endothelial Cells, Arrowheads Indicate Lymphocytes, and Asterisks Indicate Tumor Cells
A
ducer and activator of transcription 3 (STAT3) as a potential predictive factor. STAT3 is a cytoplasmic transcription factor that might translocate into the nucleus after growth factor stimulation. In contrast to normal cells in which phosphorylation is transient, constitutive activation of STAT3 has been reported in several primary cancers and tumor cell lines.10-13 Phosphorylation allows nuclear translocation and DNA binding of the STAT3 dimer, leading to upregulation of several genes involved in the cell cycle and survival such as CCND1, MYC, or BCL2L1, as well as inhibition of ligands controlling innate immunity activation.13-17 Several reports have implicated STAT3 in colon cancer oncogenesis.10-12 In particular, STAT3 was formally identified as a molecular link between chronic intestinal inflammatory diseases and CRCs.18 It has been shown that STAT3 overexpression was associated with higher CRC-specific mortality, in line with a direct role of STAT3 in mCRC prognosis.19 Another function of STAT3 in mCRC might be the activation of molecular networks leading to chemotherapy resistance. Indeed, STAT3 signaling was shown to regulate EME1 endonuclease to reduce DNA damage.20 Besides EGFR,21 several other signaling pathways such as interleukin-618 or Src kinases12 might induce STAT3 activation in cancer cells and prevent the ability of anti-EGFR to counteract STAT3related oncogenic properties. We decided to assess the clinical outcomes of anti-EGFR– containing chemotherapy regimens in patients with mCRC according to pSTAT3 status.
Patients and Methods Patient Population Ninety-four consecutive patients with histologically confirmed mCRC and available tumor material for molecular analysis, who were treated with cetuximab-containing chemotherapy in secondline treatment or beyond, at the University Hospital of Besançon (Franche-Comté, France) between July 2004 and December 2009
B
were enrolled. Tumor location, World Health Organization (WHO) performance status, stage at diagnosis, number of metastases, higher grade of skin rash during treatment, sampling method of tumors, and chemotherapy administrated with cetuximab were included in the database. KRAS status was routinely ascertained from July 2008.
Immunohistochemical Analysis Formalin-fixed and paraffin-embedded tumor samples obtained before therapy were cut in 4-m sections using standard techniques. Immunostaining for pSTAT3 (Tyr705) was performed on the automated Benchmark XT automated slide preparation system (Ventana Medical Systems, Inc, Oro Valley, AZ) using the rabbit polyclonal antibody pSTAT3 (sc-7993, dilution 1:200; Santa Cruz Biotechnology Inc, Santa Cruz, CA). Briefly, sections were deparaffinized and antigen retrieval was carried out by incubating slides for 60 minutes in Cell Conditioning 1 (CC1) buffer. Sections were then incubated for 32 minutes with the primary rabbit polyclonal antibody pSTAT3 followed by biotin-labeled secondary anti–rabbit antibody for 8 minutes, and finally a solution of streptavidin-peroxidase for 8 minutes. Tissues were then stained for 8 minutes with diaminobenzidine substrate. Some serial sections were also incubated with antibody diluent only to provide negative controls. The evaluation of pSTAT3 expression was performed at high magnification (⫻400) on the tumor material obtained just before receiving cetuximab-containing treatment (biopsy samples or resection results of primary tumor or metastasis). Five carcinomatous fields were selected to count the number of cancer cells exhibiting nuclear staining. As previously described in colorectal carcinoma, biopsy samples were considered positive for pSTAT3 if nuclear staining was detected in more than 15% of tumor cells.22 Endothelial cells, which often stain positive for pSTAT3, served as internal positive controls. Lymph nodes from inflammatory diseases were used as external positive controls to set up pSTAT3 staining. Cytoplasmic staining was not taken into account. Figure 1 illustrates positive and
Clinical Colorectal Cancer March 2013
29
STAT3 Predicts Anti-EGFR-Based Therapy Efficiency Table 1 Patient Characteristics pSTAT3 Variable
No. (%) —
ⴙ
P Value
Sex Male
50 (53.2)
41 (57.7)
9 (39.1)
Female
44 (46.8)
30 (42.3)
14 (60.9)
.15 Median Age (y) (Range)
65.6 (29-91)
Age ⱕ 70
60 (63.8)
43 (60.6)
17 (73.9)
Age ⬎ 70
34 (36.2)
28 (39.4)
6 (26.1)
0
45 (47.9)
33 (46.5)
12 (52.2)
1
38 (40.4)
31 (43.7)
7 (30.4)
2
11 (11.7)
7 (09.9)
4 (17.4)
Resection of primary tumor
31 (33.0)
24 (33.8)
7 (30.4)
Resection of metastasis
31 (33.0)
26 (36.6)
5 (21.7)
Biopsy of primary tumor
15 (16.0)
11 (15.5)
4 (17.4)
Biopsy of metastasis
17 (18.0)
10 (14.1)
7 (30.4)
Colon
59 (62.8)
48 (67.6)
11 (47.8)
Rectum
35 (37.2)
23 (32.4)
12 (52.2)
.32
ECOG Performance Status
.42
Sampling Method
.28
Tumor Location .13 Stage at Diagnosis III
19 (20.2)
15 (21.1)
4 (17.4)
III
26 (27.7)
20 (28.2)
6 (26.1)
IV
49 (52.1)
36 (50.7)
13 (56.5)
1
29 (30.9)
24 (33.8)
5 (21.7)
ⱖ2
65 (69.1)
47 (66.2)
18 (78.3)
Grade 0/1
34 (37.8)
26 (38.2)
8 (36.4)
Grade 2/3
56 (62.2)
42 (61.8)
14 (63.6)
Before 09/2005
48 (51.0)
39 (54.9)
9 (39.2)
After 10/2005
46 (49.0)
32 (45.1)
14 (60.8)
Negative
29 (30.9)
29 (40.9)
0 (0.0)
Positive
65 (69.1)
42 (59.1)
23 (100.0)
Wild type
61 (74.4)
48 (77.4)
13 (65.0)
Mutant
21 (25.6)
14 (22.6)
7 (35.0)
Second
57 (60.6)
42 (59.2)
15 (65.2)
Third
37 (39.4)
29 (40.8)
8 (34.8)
No
24 (25.5)
15 (21.1)
9 (39.1)
Yes
70 (74.5)
56 (78.9)
14 (60.9)
Irinotecan-based
84 (89.4)
62 (87.3)
22 (95.7)
Oxaliplatin-based
10 (10.6)
9 (12.7)
1 (4.3)
.87
Number of Metastases .31
Skin Toxicity .99 Date of Fixation .3 pSTAT3 Status in TILs < .01 KRAS Status .37 Cetuximab Administration Line .63 Exposure to Irinotecan Before Cetuximab .1 Chemotherapy Regimen Administrated With Cetuximab .44
Abbreviations: ECOG ⫽ Eastern Cooperative Oncology Group; pSTAT3 ⫽ phosphorylated signal transducer and activator of transcription 3; TILs ⫽ tumor infiltrating lymphocytes.
30
Clinical Colorectal Cancer March 2013
Erion Dobi et al Table 2 Evaluation of Treatment Efficacy Regarding RECIST Criteria Variable
No. (%)
Response a
OR
SD
PD
11 (32.4)
4 (14.3)
P Value
Stage at Diagnosis II
19 (20.2)
4 (12.5)
III
26 (27.7)
13 (40.6)
6 (17.6)
7 (25.0)
IV
49 (52.1)
15 (46.9)
17 (50.0)
17 (60.7)
1
29 (30.9)
10 (31.3)
13 (38.2)
6 (21.4)
ⱖ2
65 (69.1)
22 (68.7)
21 (61.8)
22 (78.6)
.10
Number of Metastases .36
Cutaneous Toxicity Grade 0/1
34 (37.8)
7 (23.3)
10 (30.3)
17 (63.0)
Grade 2/3
56 (62.2)
23 (76.7)
23 (69.7)
10 (37.0)
Negative
71 (75.5)
29 (90.6)
21 (61.8)
21 (75.0)
Positive
23 (24.5)
3 (9.4)
13 (38.2)
7 (25.0)
Negative
29 (30.9)
14 (40.7)
8 (23.5)
7 (25.9)
Positive
65 (69.1)
19 (59.3)
26 (76.5)
20 (74.1)
Wild type
61 (74.4)
27 (93.1)
19 (65.5)
15 (62.5)
Mutant
21 (25.6)
2 (6.9)
10 (34.5)
9 (37.5)
2
57 (60.6)
19 (59.4)
20 (58.8)
18 (64.3)
3
37 (39.4)
13 (40.6)
14 (41.2)
10 (35.7)
No
24 (25.5)
8 (25.0)
8 (23.5)
8 (28.6)
Yes
70 (74.5)
24 (75.0)
26 (76.5)
20 (71.4)
< .01
pSTAT3 Status in Tumor Cells .02
pSTAT3 Status in TILs .13
KRAS Status .02
Cetuximab Administration Line .89
Exposure to Irinotecan Before Cetuximab .89
Abbreviations: CR ⫽ complete response; OR ⫽ objective response; PD ⫽ progressive disease; PR ⫽ partial response; pSTAT3 ⫽ phosphorylated signal transducer and activator of transcription 3; SD ⫽ stable disease; TILs ⫽ tumor-infiltrating lymphocytes. a OR ⫽ PR ⫹ CR.
negative pSTAT3 staining. Of note, pSTAT3 expression was not affected by the length of storage (Table 1). Analysis of the immunohistochemical staining was interpreted by 2 independent pathologists. In case of discrepancies, the pathologists reviewed the slides together and reached consensus.
Multiplex PCR products were checked for quality and yield by running 18 L in 2% agarose-Tris-borate electrophoresis gel. After purification using the gel extraction kit NucleoSpin Extract II (Macherey-Nagel, Hoerdt, France), PCR products were analyzed for the presence of KRAS codon 12 and codon 13 mutations at nucleotides c.34, c.35, c.37, and c.38.
Samples and DNA Extraction and KRAS Mutation Analysis
Statistical Analysis
Formalin-fixed and paraffin-embedded tissues containing at least 60% tumor cells were selected, and DNA extraction was performed by using the QIAamp DNA Mini Kit (Qiagen, Courtaboeuf, France). Briefly, polymerase chain reaction (PCR) was performed in a 25-L reaction containing 1⫻ Qiagen Multiplex PCR Master Mix (Qiagen, Courtaboeuf, France), Plzf and Af4 primers at 0.5-M final concentration, and Tbxas and Rag1 primers at a final concentration of 0.25 M. A multiplex PCR assay was designed to amplify fragments of KRAS exon 2 (200 bp) and to identify activating mutations using a SNaPshot assay.23
Quantitative variables were recoded into categorical variables. All categorical data were compared using the Fisher exact test or 2 test. Time to progression (TTP) and overall survival (OS) were measured from the date of initiation of the first cetuximab-containing treatment line to the first recurrence or death, respectively. The endpoint date was December 31, 2010. TTP and OS were calculated with the Kaplan-Meier method. A search for parameters related to TTP and OS length was performed by univariate analysis using the log-rank test. The univariate analysis considered the following variables: sex, age (ⱕ 70 vs.
Clinical Colorectal Cancer March 2013
31
STAT3 Predicts Anti-EGFR-Based Therapy Efficiency Table 3 TTP and OS in Second-Line or Beyond Cetuximab-Containing Treatment According to Clinicopathologic Characteristics Variable
No. (%)
TTP (Mo) Median (95% CI)
P Value
OS (Mo) Median (95% CI)
P Value
Sex Male
50 (53.2)
5.3 (4.3-6.3)
Female
44 (46.8)
6.2 (5.1-7.3)
Median Age (y) (Range)
.37
11.9 (8.0-15.8) 12.0 (9.5-14.6)
.31
65.6 (29-91)
Age ⱕ 70
60 (63.8)
5.7 (4.5-6.8)
Age ⬎ 70
34 (36.2)
5.7 (4.4-6.9)
0
45 (47.9)
6.2 (4.6-7.8)
1
38 (40.4)
5.7 (2.6-8.6)
2
11 (11.7)
2.9 (1.1-6.1)
Colon
59 (62.8)
6.7 (5.2-8.1)
Rectum
35 (37.2)
5.2 (4.0-6.3)
.85
12.4 (10.7-14.1) 10.0 (5.2-14.8)
.56
ECOG Performance Status 15.0 (11.1-19.2) .08
10.7 (6.9-14.5)
.01
5.8 (1.8-9.7)
Tumor Location .12
12.1 (9.9-14.4) 11.0 (6.7-15.3)
.53
Stage at Diagnosis II
19 (20.2)
5.7 (3.2-8.1)
III
26 (27.7)
6.1 (3.0-9.2)
IV
49 (52.1)
5.6 (4.6-6.7)
1
29 (30.9)
6.2 (5.1-7.3)
ⱖ2
65 (69.1)
5.6 (4.6-6.5)
Grade 1/2
34 (37.8)
3.3 (1.0-5.6)
Grade 2/3
56 (62.2)
6.3 (5.2-7.3)
14.9 (6.4-23.4) .34
14.0 (9.4-18.8)
.15
10.0 (7.3-12.7)
Number of Metastases .50
14.5 (9.1-19.6) 11.8 (8.8-14.8)
.38
Cutaneous Toxicity < .01
5.1 (4.0-6.2) 14.8 (11.2-18.5)
< .01
pSTAT3 Status in Tumor Cells Negative
71 (75.5)
6.3 (5.1-7.6)
Positive
23 (24.5)
4.5 (4.1-4.9)
Negative
29 (31.5)
6.2 (3.4-9.0)
Positive
63 (68.5)
5.7 (4.1-7.2)
< .01
13.1 (7.6-18.7) 9.4 (7.2-11.6)
.02
pSTAT3 Status in TILs .05
14.5 (6.6-22.4) 12.0 (9.8-14.3)
.37
KRAS Status Wild type
61 (74.4)
6.6 (5.3-7.8)
Mutant
21 (25.6)
3.9 (2.6-5.2)
Second
57 (60.6)
5.6 (4.6-6.6)
Third
37 (39.4)
6.0 (4.9-7.1)
< .0001
13.5 (10.0-16.9) 8.8 (4.9-12.4)
< .01
Cetuximab Administration Line .81
12.0 (8.9-15.1) 13.0 (8.0-17.9)
.86
Exposure to Irinotecan Before Cetuximab No
24 (25.5)
5.1 (3.7-6.5)
Yes
70 (74.5)
6.1 (5.1-7.1)
.23
12.7 (4.9-20.5) 11.8 (9.2-14.4)
.51
Chemotherapy Regimen Administered With Cetuximab Irinotecan-based
84 (89.4)
5.6 (4.8-6.3)
Oxaliplatin-based
10 (10.6)
7.2 (5.5-8.9)
.95
11.0 (8.4-13.7) 21.9 (0.9-42.9)
.34
Abbreviations: CI ⫽ confidence interval; ECOG ⫽ Eastern Cooperative Oncology Group; OS ⫽ overall survival; pSTAT3 ⫽ phosphorylated signal transducer and activator of transcription 3; TILs ⫽ tumor infiltrating lymphocytes; TTP ⫽ time to progression.
32
Clinical Colorectal Cancer March 2013
Erion Dobi et al Table 4 Multivariate Analysis for Objective Response Variable
Response Rate, No. (%)
Odds Ratio (95% CI)
P Value
KRAS (Wild-Type)
27/61 (44.3%)
0.17 (0.04-0.83)
.03
PSTAT3 (Negative)
29/71 (40.8%)
0.19 (0.04-0.91)
.04
Abbreviation: pSTAT3 ⫽ phosphorylated signal transducer and activator of transcription 3.
⬎ 70 years), WHO performance status (0 vs. 1-2), tumor location (colon vs. rectum), stage at diagnosis (stage II vs. III vs. stage IV), number of metastases before cetuximab initiation (1 vs. ⱖ 2), skin rash (grades 0/1 vs. grades 2/3), nuclear pSTAT3 status in tumor cells (negative vs. positive), pSTAT3 status in tumor-infiltrating lymphocytes (negative vs. positive), KRAS status, cetuximab line of treatment (second vs. third), chemotherapy administrated with cetuximab (irinotecan-based vs. oxaliplatin-based chemotherapy) and if the patients were exposed to irinotecan before cetuximabcontaining treatment. Odds ratios were estimated for objective response by logistic regression model. Multivariate analysis was performed to identify independent prognostic factors for survival using a Cox proportional hazard model with forward stepwise selection. The significance levels for entry and for stay were 0.2 and 0.1, respectively. Statistical significance for all analyses in the present report is considered at the 2-sided 5% level. The data analysis was generated using SPSS software, version 18.0.0 (IBM Corp, Armonk, NY).
Results Patients Table 1 shows the characteristics of the 94 patients. All patients were treated with cetuximab in combination with chemotherapy (89.4% with irinotecan, 10.6% with oxaliplatin) as second-line treatment or beyond. There were no patients who received the antiEGFR treatment in the first-line setting. Fifty-seven (60.6%) patients received the treatment in the second-line setting, and the remaining 37 (39.4%) patients had it as third-line treatment. Seventy (74.5%) patients had been exposed to irinotecan before cetuximabcontaining treatment. The immunohistochemical expression of nuclear pSTAT3 in tumor cells was determined in all 94 patients with mCRC, whereas KRAS status was determined in the 82 patients for whom tumor material was available. The nuclear pSTAT3 expression and KRAS mutations were detected in 23 (24.5%) and 21 (25.6%) of the tumor samples, respectively. The median follow-up was 22 months and all patients were assessable for clinical objective response, TTP, and OS. Disease characteristics were compatible with common features of patients with mCRC. Clinical outcomes in this cohort of patients were also in agreement with previously reported results. Regarding cetuximab therapy, grade ⱖ 2 acneiform rash occurred in 62% of the patients. Overall response rate assessed by the Response Evaluation Criteria in Solid Tumors (RECIST) criteria was observed in 33% of the patients (Table 2). TTP and OS were 5.8 and 12.5 months, respectively (Table 3). Of note, classic predictive factors were significantly associated with objective response rate (ORR) and TTP in this cohort. Indeed, ORR was increased for patients with KRAS WT
tumors (44% vs. 9% in the case of KRAS mutant tumors; P ⫽ .02) and TTP was 6.6 months for these patients compared with 3.9 months for patients whose tumors were KRAS mutant. As seen in Tables 2 and 3, skin rash was also associated with ORR (41% vs. 21%) and TTP (6.3 vs. 3.3 months).
Immunohistological Analysis of pSTAT3 Status The kappa value for concordance between both pathologists for evaluating pSTAT3 status was 0.86. To control for potential bias, the impact of the sampling method on the nuclear immunohistochemical expression of pSTAT3 was monitored. There was no significant difference dependent on the tumor sampling method (P ⫽ .36). With a cutoff of 15%, we identified 23 patients (24.5%) whose tumors expressed high levels of pSTAT3 within the nucleus of tumor cells. As shown in Table 1, we did not observe any significant influence between the patients characteristics and the expression of pSTAT3. Moreover, the status of pSTAT3 did not differ in KRAS-WT or KRAS-mutant mCRC (Table 1). Interestingly, tumorinfiltrating lymphocytes (TILs) expressing pSTAT3 were observed in all 23 pSTAT3-positive tumors, whereas only 59% of pSTAT3negative tumors were associated with pSTAT3-positive TIL.
Influence of pSTAT3 Status on Tumor Response to Cetuximab-containing Chemotherapy ORR was significantly lower among patients with positive nuclear expression of pSTAT3 regarding RECIST criteria. The probability of achieving an objective response was 13% when a positive nuclear expression of pSTAT3 was reported, compared with 41% for patients displaying pSTAT3-negative tumors (P ⫽ .02) (Table 2). In the multivariate logistic regression model, pSTAT3-negative status and KRAS WT status both remained significant to predict overall responses occurring in the context of cetuximab-containing chemotherapy regimens (odds ratio [OR], 0.17; 95% confidence interval [CI], 0.04-0.83; P ⫽ .04; OR, 0.19; 95% CI, 0.04-0.91; P ⫽ .03, respectively) (Table 4).
Influence of pSTAT3 Status on Time to Progression In univariate analysis regarding TTP, patients with expression of pSTAT3 in tumor cells had a worse prognosis (4.5 vs. 6.3 months; P ⬍ .01) (Table 3; Figure 2A). Among patients with WT KRAS tumors, expression of nuclear pSTAT3 in tumor cells remained significantly associated with worse TTP (5.1 vs. 7.1 months, P ⫽ .04). pSTAT3 did not influence TTP in patients whose tumors displayed KRAS mutations (4.5 vs. 3.3 months, P ⫽ .52; Table 5). In multivariate analysis, the following factors remained significant for length of TTP: KRAS WT status (hazard ratio [HR], 0.4; 95% CI, 0.2-0.7; P ⬍ .01), negative pSTAT3 status (HR, 0.5; 95% CI, 0.3-.09; P ⫽ .02), and high-grade skin rash (HR, 0.6; 95% CI, 0.4-.09; P ⫽ .02) (Table 6).
Influence of pSTAT3 Status on Overall Survival In univariate analysis regarding OS, patients with positive nuclear expression of pSTAT3 in tumor cells had a worse OS (9.4 vs. 13.1 months; P ⫽ .02) (Figure 2B). OS was significantly lower among patients whose tumors carried KRAS mutations (8.8 vs. 13.5 months; P ⬍ .01).
Clinical Colorectal Cancer March 2013
33
STAT3 Predicts Anti-EGFR-Based Therapy Efficiency Figure 2 Kaplan-Meier Curves According to pSTAT3 Status for Time to Tumor Progression (A) and Overall Survival (B)
A 1.0
B 1.0 pSTAT3 status negative positive
0.8
0.6
Probability
Probability
0.8
pSTAT3 status negative positive
0.4
0.2
0.6
0.4
0.2 Log-rank P < 0.01
Log-rank P = 0.02
0.0
0.0 0
No. at risk Negative 71 Positive 23
6 12 18 Time to tumor progression (months)
44 8
12 0
3 0
24
0
1 0
No. at risk Negative 71 Positive 23
6 12 18 Overall survival (months)
24
52 16
18 2
37 10
25 2
Table 5 TTP and OS in Second-Line or Beyond Cetuximab-Containing Treatment According to pSTAT3 Status for KRAS WT and KRAS MT Variable
No.
TTP (mo) Median (95% CI)
P Value
OS (mo) Median (95% CI)
P Value
pSTAT3 Status in KRAS WT Tumor Cells Negative
48
7.1 (5.6-8.5)
Positive
13
5.1 (3.7-6.5)
Negative
14
3.3 (1.5-5.1)
Positive
7
4.5 (4.2-4.9)
.04
16.6 (9.8-23.4) 12.7 (7.3-13.2)
.13
pSTAT3 Status in KRAS MT Tumor Cells .52
7.3 (0.3-14.4) 8.8 (0.2-17.4)
.34
Abbreviations: MT ⫽ mutant; OS ⫽ overall survival; pSTAT3 ⫽ phosphorylated signal transducer and activator of transcription 3; TTP ⫽ time to progression; WT ⫽ wild type.
In multivariate analysis regarding OS, the independent significant prognostic factors were high-grade skin rash (HR, 0.5; 95% CI, 0.3-0.8; P ⬍ .01), KRAS WT status (HR, 0.5; 95% CI, 0.3-0.9; P ⫽ .02), and negative pSTAT3 status (HR, 0.5; 95% CI, 0.3-0.9; P ⫽ .03).
Discussion The addition of anti-EGFR monoclonal antibodies to conventional therapies extended the therapeutic strategies available in mCRC management. EGFR targeting in this disease has been demonstrated as a relevant option to improve overall response along with to standard chemotherapy regimens3,24 or to restore the efficacy of irinotecan in previously treated patients with mCRC.25 Although the presence of KRAS mutations is a specific predictive biomarker for lack of anti-EGFR monoclonal antibody efficacy, there is convincing evidence that additional events are involved in this process, since not all patients with KRAS WT tumors benefit from anti-EGFR mono-
34
Clinical Colorectal Cancer March 2013
clonal antibodies.25,26 We have demonstrated, for the first time, that activated STAT3 (pSTAT3) may contribute, along with KRAS status, to the selection of patients eligible for anti-EGFR– based therapies. Patient characteristics and clinical outcomes observed in our cohort are consistent with previously published literature and with patients treated in general practice. KRAS molecular status and skin rash were correlated to cetuximab efficacy. In the present study, 24.5% of the tumor samples displayed a high nuclear expression of pSTAT3. This observation is in agreement with the 18% level of pSTAT3 expression reported recently in mCRC.19 The negative impact of pSTAT3 on clinical outcomes achieved by anti-EGFR therapy is consistent with the pivotal role of STAT3 signaling in oncogenesis.27 Several mechanisms driven by the transcriptional activity of STAT3 might promote cancer progression.
Erion Dobi et al Table 6 Multivariate Analysis for TTP and OS Variable
HR (95% CI)
P Value
KRAS (wild type)
0.4 (0.2-0.7)
< .01
Skin rash (grade 2/3)
0.6 (0.4-0.9)
.02
pSTAT3 (negative)
0.5 (0.3-0.9)
.02
Skin rash (grade 2/3)
0.5 (0.3-0.8)
< .01
KRAS (wild type)
0.5 (0.3-0.9)
.02
pSTAT3 (negative)
0.5 (0.3-0.9)
.03
●
TTP ● ●
OS
Abbreviations: OS ⫽ overall survival; pSTAT3 ⫽ phosphorylated signal transducer and activator of transcription 3; TTP ⫽ time to progression.
Indeed, STAT3 directly induces cyclin D1 and vascular endothelial growth factor transcription in tumor cells, leading to cancer proliferation and angiogenesis.15,28 The adverse prognostic value of STAT3 was recently confirmed in patients with mCRC. Morikawa et al established a negative correlation between pSTAT3 expression in CRC cells and patient survival.19 A possible hypothesis accounting for the predictive value of STAT3 might be the ability of this signaling pathway to confer resistance to chemotherapy. In vitro, STAT3 activation was shown to be related to chemotherapy resistance in B-cell lymphoma,29 multiple myeloma,30 and CRC.21 STAT3 transcriptional activity increases BCL2L1 and MCL1 expression, leading to drug resistance by inhibiting the cell death pathways associated with genotoxic treatment. Other experiments involved the interleukin-6/Jak2/STAT3 pathway in cancer stem cell homeostasis, which had been shown to include most of the chemotherapy-resistant cells.31-33 In the clinical setting, the presence of pSTAT3 in breast cancer cells is correlated to a lower probability of complete pathologic response after neoadjuvant chemotherapy, sustaining a possible role for pSTAT3 in chemotherapy resistance.34 The potential involvement of STAT3 signaling in chemotherapy resistance is illustrated in our study by a decreased ORR achieved by anti-EGFR and chemotherapy when pSTAT3 status was observed in cancer cells (Table 2). These exploratory results prompt the development of further clinical studies to confirm the predictive value of pSTAT3 on anti-EGFR therapy effectiveness. It would be of particular interest to confirm the clinical relevance of these results by comparing the impact of pSTAT3 in a controlled randomized trial assessing the efficacy of chemotherapy with or without anti-EGFR. Furthermore, the role of STAT3 as a molecular target to optimize anti-EGFR therapies was recently highlighted.35 A preclinical study demonstrated that in contrast to cetuximab monotherapy, dual inhibition of EGFR signaling using cetuximab and erlotinib inhibits pSTAT3.35 Inhibition of STAT3 in mCRCs might be a promising strategy to target chemotherapy-resistant cancer cells and should be assessed in association with anti-EGFR monoclonal antibodies.
Clinical Practice Points ●
Targeting the extracellular domain of EGFR with monoclonal antibodies (cetuximab or panitumumab) is an effective strategy as
●
● ●
a single agent or in combination with chemotherapy regimens for patients with mCRC and KRAS WT tumors. Currently, KRAS codon 12 and codon 13 mutations are the only predictive biomarkers used in clinical practice to prescribe antiEGFR antibodies. In this retrospective study, we observed an activation of STAT3 in 24.5% of the tumor samples analyzed. Our results showed a negative impact of pSTAT3 in the ORR achieved by an anti-EGFR and chemotherapy regimen combination and decreased progression-free and OS in patients with mCRC displaying an active form of STAT3 in their tumors. Immunohistochemical analysis of pSTAT3 status might be of clinical interest to better individualize the potential benefit of antiEGFR therapies. The predictive value of this biomarker should be confirmed in further randomized clinical trials. Moreover, these results sustain the development of preclinical and clinical research to assess the impact of STAT3 molecular inhibition to potentiate anti-EGFR therapies.
Acknowledgments We are grateful to Ms Alice Baraquin, Ms Marthe Bigand, Ms Adeline Bouard, Mr Christophe Bracieux, Mr Benoît Girardo, and Ms Monique Guyot from the Department of Molecular Biology and Pathology for their excellent technical assistance with tumor section preparation and immunohistochemical and biological analysis. We also thank the biobank Tumorothèque Régionale de Franche-Comté for making available all useful tumor samples for the research. This work was supported by a research grant from the “Ligue contre le cancer du Grand Est”.
Disclosure The authors have stated that they have no conflicts of interest.
References 1. Hurwitz H, Fehrenbacher L, Novotny W, et al. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 2004; 350:2335-42. 2. Van Cutsem E, Köhne CH, Láng I, et al. Cetuximab plus irinotecan, fluorouracil, and leucovorin as first-line treatment for metastatic colorectal cancer: updated analysis of overall survival according to tumor KRAS and BRAF mutation status. J Clin Oncol 2011; 29:2011-9. 3. Douillard JY, Siena S, Cassidy J, et al. Randomized, phase III trial of panitumumab with infusional fluorouracil, leucovorin, and oxaliplatin (FOLFOX4) versus FOLFOX4 alone as first-line treatment in patients with previously untreated metastatic colorectal cancer: the PRIME study. J Clin Oncol 2010; 28:4697-705. 4. Van Cutsem E, Peeters M, Siena S, et al. Open-label phase III trial of panitumumab plus best supportive care compared with best supportive care alone in patients with chemotherapy-refractory metastatic colorectal cancer. J Clin Oncol 2007; 25:1658-64. 5. Lièvre A, Bachet JB, Le Corre D, et al. KRAS mutation status is predictive of response to cetuximab therapy in colorectal cancer. Cancer Res 2006; 66:3992-5. 6. Di Fiore F, Blanchard F, Charbonnier F, et al. Clinical relevance of KRAS mutation detection in metastatic colorectal cancer treated by cetuximab plus chemotherapy. Br J Cancer 2007; 96:1166-9. 7. Karapetis CS, Khambata-Ford S, Jonker DJ, et al. K-ras mutations and benefit from cetuximab in advanced colorectal cancer. N Engl J Med 2008; 359:1757-65. 8. Amado RG, Wolf M, Peeters M, et al. Wild-type KRAS is required for panitumumab efficacy in patients with metastatic colorectal cancer. J Clin Oncol 2008; 26:1626-34. 9. Yu H, Jove H. The STATs of cancer—new molecular targets come of age. Nat Rev Cancer 2004; 4:97-105. 10. Levy DE, Lee CK. What does Stat3 do? J Clin Invest 2002; 109:1143-8. 11. Bromberg JF, Horvath CM, Besser D, et al. Stat3 activation is required for cellular transformation by v-src. Mol Cell Biol 1998; 18:2553-8.
Clinical Colorectal Cancer March 2013
35
STAT3 Predicts Anti-EGFR-Based Therapy Efficiency 12. Bedel R, Thiery-Vuillemin A, Grandclement C, et al. Novel role for STAT3 in transcriptional regulation of NK immune cell targeting receptor MICA on cancer cells. Cancer Res 2011; 71:1615-26. 13. Barré B, Vigneron A, Coqueret O. The STAT3 transcription factor is a target for the Myc and riboblastoma proteins on the Cdc25A promoter. J Biol Chem 2005; 280: 15673-81. 14. Leslie K, Lang C, Devgan G, et al. Cyclin D1 is transcriptionally regulated by and required for transformation by activated signal transducer and activator of transcription 3. Cancer Res 2006; 66:2544-52. 15. Shirogane T, Fukada T, Muller JM, et al. Synergistic roles for Pim-1 and c-Myc in STAT3-mediated cell cycle progression and antiapoptosis. Immunity 1999; 11:709-19. 16. Kusaba T, Nakayama T, Yamazumi K, et al. Activation of STAT3 is a marker of poor prognosis in human colorectal cancer. Oncol Rep 2006; 15:1445-51. 17. Bollrath J, Phesse TJ, von Burstin VA, et al. gp130-mediated Stat3 activation in enterocytes regulates cell survival and cell-cycle progression during colitis-associated tumorigenesis. Cancer Cell 2009; 15:91-102. 18. Morikawa T, Baba Y, Yamauchi M, et al. STAT3 expression, molecular features, inflammation patterns, and prognosis in a database of 724 colorectal cancers. Clin Cancer Res 2011; 17:1452-62. 19. Vigneron A, Gamelin E, Coqueret O. The EGFR-STAT3 oncogenic pathway upregulates the Eme1 endonuclease to reduce DNA damage after topoisomerase I inhibition. Cancer Res 2008; 68:815-25. 20. Grandis JR, Drenning SD, Chakraborty A, et al. Requirement of Stat3 but not Stat1 activation for epidermal growth factor receptor–mediated cell growth in vitro. J Clin Invest 1998; 102:1385-92. 21. Barré B, Vigneron A, Perkins N, et al. The STAT3 oncogene as a predictive marker of drug resistance. Trends Mol Med 2007; 13:4-11. 22. Magnin S, Viel E, Baraquin A, et al. A multiplex SNaPshot assay as a rapid method for detecting KRAS and BRAF mutations in advanced colorectal cancers. J Mol Diagn 2011; 13:485-92. 23. Bokemeyer C, Bondarenko I, Hartmann JT, et al. Efficacy according to biomarker status of cetuximab plus FOLFOX-4 as first-line treatment for metastatic colorectal cancer: the OPUS study. Ann Oncol 2011; 22:1535-46.
36
Clinical Colorectal Cancer March 2013
24. Cunningham D, Humblet Y, Siena S, et al. Cetuximab monotherapy and cetuximab plus irinotecan in irinotecan-refractory metastatic colorectal cancer. N Engl J Med 2004; 351:337-45. 25. Wong R, Cunningham D. Using predictive biomarkers to select patients with advanced colorectal cancer for treatment with epidermal growth factor receptor antibodies. J Clin Oncol 2008; 26:5668-70. 26. Bromberg JF, Wrzeszczynska MH, Devgan G, et al. Stat3 as an oncogene. Cell 1999; 98:295-303. 27. Niu G, Wright KL, Huang M, et al. Constitutive Stat3 activity up-regulates VEGF expression and tumor angiogenesis. Oncogene 2002; 21:2000-8. 28. Alas S, Bonavida B. Rituximab inactivates signal transducer and activation of transcription 3 (STAT3) activity in B-non-Hodgkin’s lymphoma through inhibition of the interleukin 10 autocrine/paracrine loop and results in down-regulation of Bcl-2 and sensitization to cytotoxic drugs. Cancer Res 2001; 61:5137-44. 29. Alas S, Bonavida B. Inhibition of constitutive STAT3 activity sensitizes resistant non-Hodgkin’s lymphoma and multiple myeloma to chemotherapeutic drug-mediated apoptosis. Clin Cancer Res 2003; 9:316-26. 30. Marotta LL, Almendro V, Marusyk A, et al. The JAK2/STAT3 signaling pathway is required for growth of CD44⫹CD24⫺ stem cell-like breast cancer cells in human tumors. J Clin Invest 2011; 121:2723-35. 31. Creighton CJ, Li X, Landis M, et al. Residual breast cancers after conventional therapy display mesenchymal as well as tumor-initiating features. Proc Natl Acad Sci U S A 2009; 106:13820-5. 32. Li X, Lewis MT, Huang J, et al. Intrinsic resistance of tumorigenic breast cancer cells to chemotherapy. J Natl Cancer Inst 2008; 100:672-9. 33. Diaz N, Minton S, Cox C, et al. Activation of STAT3 in primary tumors from high-risk breast cancer patients is associated with elevated levels of activated SRC and survivin expression. Clin Cancer Res 2006; 12:20-8. 34. Ruff-Jamison S, Zhong Z, Wen Z, et al. Epidermal growth factor and lipopolysaccharide activate Stat3 transcription factor in mouse liver. J Biol Chem 1994; 269: 21933-45. 35. Weickhardt AJ, Price TJ, Chong G, et al. Dual targeting of the epidermal growth factor receptor using the combination of cetuximab and erlotinib: preclinical evaluation and results of the phase II DUX study in chemotherapy-refractory, advanced colorectal cancer. J Clin Oncol 2012; 30:1505-12.