- Email: [email protected]
EGFR FISH assay predicts for response to cetuximab in chemotherapy refractory colorectal cancer patients
original article
Annals of Oncology 19: 717–723, 2008 doi:10.1093/annonc/mdm492 Published online 31 December 2007
EGFR FISH assay predicts for response to cetuximab in chemotherapy refractory colorectal cancer patients F. Cappuzzo1*, G. Finocchiaro1,2, E. Rossi3, P.A. Ja¨nne4, C. Carnaghi5, C. Calandri6, K. Bencardino7, C. Ligorio1, F. Ciardiello7, T. Pressiani1, A. Destro1,8, M. Roncalli1,8, L. Crino9, W. A. Franklin2, A. Santoro1 & M. Varella-Garcia1 1
Background: Standardized conditions to distinguish subpopulations of colorectal cancer (CRC) patients more and less sensitive to cetuximab therapy remain undefined.
Materials and methods: We retrospectively analyzed epidermal growth factor receptor (EGFR) copy number by fluorescence in situ hybridization (FISH) in paraffin-embedded tumor blocks from 85 chemorefractory CRC patients treated with cetuximab. Results were analyzed according to different score systems previously reported in colorectal and lung cancers. The primary end point of the study was identification of the EGFR FISH score that best associates with response rate (RR). Results: Using receiver operating characteristic (ROC) analysis, the cut-off that best discriminated responders versus nonresponders to cetuximab was a mean of 2.92 EGFR gene copies per cell. This model showed sensitivity of 58.6% [95% confidence interval (CI) = 47.1–70.1) and specificity of 93.3% (95% CI = 80.6–100). EGFR FISH-positive patients (N = 43, 50.6%) had significantly higher RR (P = 0.0001) and significantly longer time to disease progression (P = 0.02) than EGFR FISH negative (N = 42, 49.4%). Other scoring systems resulted less accurate in discriminating patients with the highest likelihood of response to cetuximab therapy. Conclusions: CRC patients with high EGFR gene copy number have an increased likelihood to respond to cetuximab therapy. Prospective clinical trials with a careful standardization of assay conditions and pattern interpretation are urgently needed. Key words: cetuximab, colon cancer, EGFR, fluorescence in situ hybridization
introduction Colorectal cancer (CRC) is one of the most common human malignancies and one of the leading causes of cancer-related death in the Western world [1]. During the last years, several strategies interfering with the epidermal growth factor receptor (EGFR) have been evaluated in solid tumors, including CRC. Although EGFR tyrosine kinase inhibitors (TKIs) demonstrated to induce dramatic and durable responses in certain patients with advanced non-small-cell lung cancer (NSCLC), no activity has been observed in CRC [2–4]. Conversely, monoclonal antibodies (mAbs) against EGFR, such as cetuximab (C225, Erbitux; ImClone System Inc, New York, NY) and panitumumab showed activity in metastatic CRC when used alone or in combination with irinotecan [5, 6]. *Correspondence to: Dr F. Cappuzzo, Istituto Clinico Humanitas, via Manzoni 56, 20089-Rozzano, Italy. Tel: +39-02-82244097; Fax: +39-02-82244097; E-mail: [email protected]
Biological mechanisms underlying response or resistance to antibodies against EGFR are only partially known. Although at the present time regulatory agencies require selections of patients for such therapy based on EGFR expression by immunohistochemistry (IHC), responses were also noted in patients whose tumors had no EGFR immunostaining [7], confirming the pitfalls of IHC in assessing EGFR expression and its limits for patient selection. In NSCLC, activating EGFR gene mutations and EGFR gene increased copy number are considered the most critical factors associated with response to TKI therapy, while KRAS mutations are responsible for intrinsic resistance to TKI in some patients [8–12]. In CRC, mutations in the tyrosine kinase domain of the EGFR gene occur at very low frequency [13] and their role in cetuximab sensitivity seem to be irrelevant [14, 15], while RAS and RAF mutations are associated with resistance to therapy with antibodies against EGFR [16–18].
ª The Author 2007. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For Permissions, please email: [email protected]
original article
Received 10 August 2007; revised 18 September 2007; accepted 19 September 2007
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on April 30, 2012
Department of Medical Oncology, Istituto Clinico Humanitas IRCCS, Rozzano, Italy; 2Department of Medicine/Medical Oncology, University of Colorado Cancer Center, Aurora, CO, USA; 3CINECA-Interuniversity Consortium, Bologna, Italy; 4Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA, USA; 5 Department of Medical Oncology, Bellaria-Maggiore Hospital, Bologna; 6Department of Medical Oncology, San Matteo Hospital, Pavia; 7Department of Medical Oncology, Seconda Universita’ di Napoli, Napoli; 8Pathology Unit, Milan University, Istituto Clinico Humanitas IRCCS, Rozzano; 9Department of Medical Oncology, Ospedale Silvestrini, Perugia, Italy
original article
materials and methods patient selection This retrospective study was conducted in a cohort of 85 CRC patients treated with cetuximab-based therapy for metastatic disease from January 2002 to August 2006 in five Italian institutions: Istituto Clinico Humanitas, Rozzano; Ospedale Bellaria, Bologna; Policlinico S.Matteo, Pavia; Ospedale Silvestrini, Perugia; and Seconda Universita` di Napoli, Napoli. For initial treatment in their institutions, patients were selected by presence of EGFR expression assessed in the local hospitals, using a commercially available kit (DAKO, Carpinteria, CA), and considering as positive all tumors with at least 1% immunostaining in tumor cells. For this study, patients were selected based on two main criteria: presence of at least one measurable lesion and availability of tumor tissue. No other clinical or biological criterion was used for patient selection. Cetuximab was given to each patient at the initial dose of 400 mg/m2, followed by weekly infusion of 250 mg/m2. In all patients, disease assessment was carried out every 2 months, with a confirmatory evaluation no <4 weeks after the response assessment, according to the response evaluation criteria in solid tumors (RECIST) criteria [22]. The study was approved by the local Ethics Committee and was conducted in accordance with ethical principles stated in the most recent version of the Declaration of Helsinki.
tissue preparation, FISH and IHC IHC and FISH assays were carried out in sections cut at 4 lm from paraffin-embedded tissue blocks containing representative malignant cells
718 | Cappuzzo et al.
and obtained at time of diagnosis. All IHC and FISH analyses were carried out at the University of Colorado Cancer Center, USA. One slide was hematoxylin–eosin (H&E) stained to be used as reference to select dominant tumor foci and sequential sections were assayed for the selected markers. The EGFR FISH assay was carried out with the LSI EGFR SpectrumOrange/CEP 7 SpectrumGreen probe (Vysis/Abbott Molecular, Des plaines, IL) according to methods described elsewhere [9, 23]. Four physically distant tumor areas were selected guided by the H&E-stained slide and the EGFR and CEP7 signals were counted in at least 30 nuclei per tumor area at ·1000 magnification. Microscope analysis was carried out independently by two observers (GF, MVG) blinded to the patients’ clinical characteristics and in case of discordance a consensus was reached in further analyses. In cases of heterogeneity among the tumor areas, only the area with the worst score was considered. For patients with tissue from primary tumor and corresponding metastasis, the sample from primary lesion was used to represent the patient in the correlative analyses with clinical outcome. Evaluation of EGFR protein expression by IHC was repeated for all specimens using a mAb against the external domain of the EGFR (Zymed Laboratories, San Francisco, CA). This analysis was carried out blindly to patients’ clinical characteristics and FISH results (by WAF) and tumor was considered positive when staining was observed in at least 1% of cells.
statistical analyses The primary end point of the study was the identification of the EGFR FISH score that best associates with response rate (RR). Secondary end points were association with time to disease progression (TTP), overall survival (OS) and concordance of EGFR profile between primary tumors and corresponding metastases. TTP was calculated from the time of first cetuximab infusion to time of disease progression or last disease assessment. OS was calculated from the time of first cetuximab infusion to patient death or last contact. Differences in RR were compared by Fisher’s exact test or v2 test. TTP, OS and the 95% confidence intervals (CIs) were evaluated by survival analysis using Kaplan–Meier method [24]. TTP and OS for the groups with negative and positive biomarker were compared using the log-rank test. Concordance between biological data in primary tumor and corresponding metastasis was calculated by weighted k statistic [25]. Statistical significance was set at <0.05 for each analysis. Analysis of receiver operating characteristic (ROC) curve was carried out with the aim of determining a cut-off point for EGFR gene copy number as a continuous variable [26]. EGFR sensitivity and specificity were expressed in terms of percentage and the value for which sensitivity and sensibility were the highest has been chosen as the best cut-off point. In addition to the score identified in the present study by ROC analysis and indicated as score A, tumors were classified as FISH positive or negative based on classifications previously reported in CRC by Sartore-Bianchi et al. [21] (score B, EGFR FISH positive = mean of ‡2.47 gene signals per nucleus and score C, CEP 7 polysomy positive = ‡43% of cells with ‡3 copies of CEP7) and in lung carcinomas [9, 23] by our group (score D, EGFR FISH positive = ‡40% of cells with ‡4 copies of EGFR or gene amplification). All statistical analyses were carried out using SPSS version 11.5.1 (SPSS Italia srl, Bologna, Italy).
results patient characteristics FISH analysis was successful in tumor samples from 85 chemorefractory patients. As shown in Table 1, our cohort presented clinical characteristics and outcome similar to CRC patients included in a large registration trial with cetuximab [5]
Volume 19 | No. 4 | April 2008
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on April 30, 2012
Recent studies indicated that EGFR gene copy number could influence response to anti-EGFR mAbs cetuximab or panitumumab therapy in CRC [18–21], although different techniques and scoring systems were adopted in each study, making direct comparison and assessment of clinical applicability practically impossible. In the study conducted by Lievre et al. [18], an increased EGFR gene copy number assessed by chromogenic in situ hybridization was significantly associated with an objective tumor response to cetuximab. However, the low number of EGFR-positive patients (three cases) precluded any firm conclusion. In the study carried out by Lenz et al. [19], EGFR gene gain assessed by PCR was significantly associated with longer survival, raising the question whether EGFR genomic gain represents a favorable prognostic factor. In the retrospective analysis conducted by Moroni et al. [20] where EGFR genomic gain was assessed by fluorescence in situ hybridization (FISH), there was a significant association between high EGFR copy numbers and better response to anti-EGFR mAbs cetuximab or panitumumab. In this study, in which patients were selected by clinical outcome (10 responders, 21 nonresponders), there was only one responder with <3 copies of EGFR per cell and only one nonresponder with >2.1 copies of EGFR per cell. More recently, the same group analyzed retrospectively 92 CRC patients treated with panitumumab plus best supportive care (BSC) or BSC alone and concluded that tumors with <2.47 EGFR copies per cell or <43% cells with chromosome 7 polysomy (‡3 chromosome enumeration probe 7 copies per cell) had a reduced likelihood of response to the therapy [21]. Because there is an urgent need for defining a method for selection of patients for anti-EGFR therapies, we investigated the association between EGFR gene copy numbers evaluated by FISH with response to cetuximab therapy in advanced CRC.
Annals of Oncology
original article
Annals of Oncology
Table 1. Patient characteristics Patient characteristics
Table 2. Biological characteristics Number
100 63.5 36.5
65.9 30.6 3.5 76.5 23.5 74.1 52.9 21.2 18.8 25.9 28.2 45.9 21.2 40 35.3 83.5 84.7 90.6 3.5 5.9
and all patients had received at least one chemotherapy line for metastatic disease. In the whole population, overall response rate was 17.7%, including one complete (1.2%) and 14 partial responses (16.5%). Median time to progression (TTP) was 5.4 months and median OS was 10.5 months. In the univariate analysis, no clinical characteristic was associated with response to cetuximab therapy. EGFR analysis was carried out in tissue from primary tumor only in 43 cases (50.6%), from metastasis only in 20 cases (23.5%) and from both primary tumor and corresponding metastasis in 22 cases (25.9%). Primary tumor tissues originated from colon in 48 cases and from rectum in 17 cases, while metastatic tissues were obtained from liver (25 cases), lung (eight cases) or other sites (nine cases).
EGFR analyses Patients were evaluated for EGFR protein expression by IHC and gene copy number by FISH (Table 2). EGFR IHC analysis was successfully carried out in 82 patients. Although all patients included in the present study had received cetuximab-based therapy, because were considered EGFR IHC positive, when tested in the local hospital using standard DAKO immunostaining, the analysis carried out in our central laboratory detected 16 patients lacking EGFR expression
Volume 19 | No. 4 | April 2008
Marker
Positive result (N/%)
EGFR IHC 66/80.5 (N = 82) EGFR FISH (N = 85) Score A 43/50.6 Score B 59/69.4 Score C 41/48.2 Score D 34/40.0
Negative result (N/%)
Reference
16/19.5
42/49.4 26/30.6 44/51.8 51/60.0
Present article Sartore-Bianchi et al. [21] Sartore-Bianchi et al. [21] Cappuzzo et al. [9], Varella-Garcia [23]
In the present study, using ROC analysis, a mean of 2.92 EGFR gene copy number qualified the tumor as FISH positive (score A). As defined by Sartori-Bianchi [21] in CRC, a mean of 2.47 EGFR gene copy number (score B) or chromosome 7 polysomy in >43% of cells (score C) qualified the tumor as FISH positive. According to the score system proposed in NSCLC, a tumor was defined as FISH positive when ‡40% of cells had ‡4 copies of EGFR or in presence of gene amplification (score D, [9, 23]). EGFR, epidermal growth factor receptor; IHC, immunohistochemistry; FISH, fluorescence in situ hybridization; ROC, receiver operating characteristic; CRC, colorectal cancer; NSCLC, non-small-cell lung cancer.
(EGFR IHC negative, Table 2). EGFR IHC-positive status was not associated with any clinical characteristic or end point (Table 3). Among the 85 patients, two (2.4%) had EGFR gene amplification in the tumor cells, while all others had balanced EGFR and CEP7 copy numbers. Thirty-four tumors had high level of genomic gain (mean EGFR = 4.2, ‡4 copies in ‡40% of nuclei) and 21 tumors had no or very low genomic gain for both tested targets (mean EGFR = 2.2). ROC analysis identified the mean of 2.92 EGFR copies per cell as the best cut-off value associated with patient outcome (Figure 1). Using this cut-off, 43 patients (50.6%) were classified as EGFR FISH positive. RR (32.6% versus 2.4%, P = 0.0001), TTP (6.6 versus 3.5 months, P = 0.02) and OS (11.3 versus 8.5 months, P = 0.8) favored EGFR FISH-positive individuals (Figure 2, Table 3). This model showed a sensitivity of 58.6% (95% CI = 47.1–70.1) and a specificity of 93.3% (95% CI = 80.6–100). Further, as presented in Table 3, our patient population was analyzed according to other scoring systems previously reported in colorectal carcinomas (scores B and C, [21]), and in lung carcinomas (score D, [9, 23]). Using score B [21], 59 patients (69.4%) were FISH positive. EGFR FISH positive had a significantly higher RR (23.7% versus 3.8%, P = 0.03), a trend towards longer TTP (6.1 versus 3.4 months, P = 0.06), with no difference in survival (9.8 versus 10.5 months, P = 0.6). Score B showed a 35.7% sensitivity (95% CI = 24.5–46.9) and 93.3% specificity (95% CI = 80.6–100). Using score C [21], 41 patients (48.2%) had chromosome 7 polysomy and were considered as FISH positive. No difference in RR (15.9% versus 19.5%), TTP (4.1 versus 6.1 months, P = 0.7) and survival (10.8 versus 9.5 months, P = 0.3) was observed between FISH negative and FISH positive, respectively. For this model, sensitivity was 52.9% (95% CI = 41.2–64.6) and specificity was
doi:10.1093/annonc/mdm492 | 719
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on April 30, 2012
Total evaluated 85 Gender Male 54 Female 31 Median age (year, range) 63.2 (29–79) Eastern Cooperative Oncology Group performance status 0 56 1 26 2 3 Primary tumor site Colon 65 Rectum 20 Metastatic sites Liver 63 Lung 45 Peritoneum 18 Other 16 Number of metastatic sites 1 22 2 24 ‡3 39 Previous chemotherapy lines 1 18 2 34 ‡3 30 Previous irinotecan 71 Previous oxaliplatin 72 Cetuximab + irinotecan 77 Cetuximab + oxaliplatin 3 Cetuximab alone 5
%
original article
Annals of Oncology
Table 3. Clinical outcome according to biological characteristics Patients (N/%)
Overall response (N/%)
Time to progression (months)
Overall survival (months)
Sensitivity (CI 95%)
Specificity (CI 95%)
EGFR IHC+ EGFR IHC2 P value EGFR FISH+ (score A) EGFR FISH– (score A) P value EGFR FISH+ (score B) EGFR FISH– (score B) P value CEP7 FISH+ (score C) CEP7 FISH– (score C) P value EGFR FISH+ (score D) EGFR FISH– (score D) P value
66/80.5 16/19.5
11/16.7 3/18.8 1.0 14/32.6 1/2.4 0.0001 14/23.7 1/3.8 0.03 8/19.5 7/15.9 0.6 8/23.5 7/13.7 0.2
4.6 6.6 0.2 6.6 3.5 0.02 6.1 3.4 0.06 6.1 4.1 0.7 6.5 4.1 0.3
9.2 13.7 0.3 11.3 8.5 0.8 9.8 10.5 0.6 9.5 10.8 0.3 9.5 10.8 0.2
19.1 (9.8–28.4)
78.6 (57.1–100)
58.6 (47.1–70.1)
93.3 (80.6–100)
35.7 (24.5–46.7)
93.3 (80.6–100)
52.9 (41.2–64.6)
53.3 (28.1–78.5)
62.9 (51.6–74.2)
53.3 (28.1–78.5)
43/50.6 42/49.4 59/69.4 26/30.6 41/48.2 44/51.8 34/40.0 51/60.0
Score A: ROC analysis; Scores B and C: Sartore-Bianchi et al. [21]; Score D: Cappuzzo et al. [9], Varella-Garcia [23]. CI, confidence interval; EGFR, epidermal growth factor receptor; IHC, immunohistochemistry; FISH, fluorescence in situ hybridization; ROC, receiver operating characteristic.
1,0
concordance in EGFR expression (IHC) was observed between primary tumor and corresponding metastasis (P = 0.08). As shown in Table 4, among the 17 primary tumors positive for EGFR by IHC, 16 were also positive in corresponding metastasis and one was negative. Among the five primary tumors with negative EGFR IHC score, four resulted positive in the corresponding metastasis. The EGFR gene copy results by FISH were similar in primary tumor and metastasis, with a significant concordance independently of the score system adopted (P = 0.001, Table 4). Using the cut-off identified by the ROC analysis, 21 patients had the same EGFR gene score, and only in one case the tumor was classified as FISH positive in the primary lesion and FISH negative in the corresponding liver metastasis.
AUC=0.76
Sensitivity
,8
,5
,3
0,0 0,0
,3
,5
,8
1,0
Specificity Figure 1. Receiver operating characteristic analysis based on mean EGFR gene copy number with response to cetuximab therapy as end point. In this model, sensitivity was 58.6% and specificity was 93.3%.
53.3% (28.1–78.5). Using score D [9, 23], 34 patients (40.0%) were classified as EGFR FISH positive and no association was detected between FISH positive and clinical characteristic or outcome such as RR (P = 0.2), TTP (p = 0.3) or OS (P = 0.2).
biomarker results in primitive tumor and corresponding metastasis In order to investigate whether EGFR status differed from primary tumor and corresponding metastases, we independently assessed paired samples from 22 patients. No
720 | Cappuzzo et al.
discussion Anti-EGFR strategies are offering new hope for patients with cancer, as demonstrated by the survival improvement obtained in patients with lung [3], breast [27], head and neck [28], pancreatic [29] and CRC [5]. However, identification of molecular predictors of anti-EGFR therapies remains a challenge. Predictive factors for sensitivity or resistance to EGFR-TKIs have been identified in lung cancer, but scarce and conflicting data exist in other diseases, including gastrointestinal cancers. In the present translational study, one of the largest so far conducted in CRC exposed to cetuximab, we showed that EGFR genomic gain was associated with response to the drug, and a mean of 2.92 EGFR gene copy number best discriminated the sensitive population. Using this cut-off value, the positive predictive value was 97.6% and the negative predictive value was 32.5%, with about 3-month gain in TTP and survival, suggesting that this patient stratum is particularly suitable for anti-EGFR therapy.
Volume 19 | No. 4 | April 2008
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on April 30, 2012
Marker
original article 1,0
1,0 ,8
Cumulative Survival Function
Cumulative Distribution Function
Annals of Oncology
p=0.02
,6 EGFR FISH+
,4
,2 0,0
EGFR FISH-
0
10
20
Time to Progression (months)
,8
p=0.8
,6 EGFR FISH+
,4 EGFR FISH-
,2 0,0
0
10
20
30
Survival (months)
Table 4. Biomarker results in primary tumor and corresponding metastasis from 22 patients Marker
Metastasis
EGFR IHC+ EGFR IHC2 EGFR FISH+ (Score A) EGFR FISH2 (Score A) EGFR FISH+ (Score B) EGFR FISH2 (Score B) EGFR FISH+ (Score C) EGFR FISH2 (Score C) EGFR FISH+ (Score D) EGFR FISH2 (Score D)
Positive 16 4 12 0 14 1 14 1 10 1
Primary
Negative 1 1 1 9 0 7 1 6 2 9
K P value 0.08 0.001 0.001 0.001 0.001
Score A: ROC analysis; Scores B and C: Sartore-Bianchi et al. [21]; Score D: Cappuzzo et al. [9], Varella-Garcia [23]. EGFR, epidermal growth factor receptor; IHC, immunohistochemistry; FISH, fluorescence in situ hybridization; ROC, receiver operating characteristic.
Present results are confirmatory of previous studies in CRC that showed that EGFR genomic gain assessed by FISH was associated with response to the mAbs cetuximab or panitumumab [20, 21]. In the study carried out by Moroni et al. where 31 patients with metastatic CRC treated with cetuximab or panitumumab were evaluated for EGFR by FISH, increased EGFR gene copy number was observed in eight of nine patients with objective response and in only one of 21 nonresponders. Nevertheless, no cut-off for discriminating EGFR FISH positive versus negative was proposed precluding attempts to reproduce these results [20]. More recently, in order to define the best cut-off discriminating responders versus nonresponders, the same group analyzed by FISH a cohort of 92 CRC patients, including 58 cases treated with panitumumab and 34 treated with BSC alone [21]. In that study, a cut-off of 2.47 mean EGFR gene copy number best discriminated nonresponders, and none of the patients with
Volume 19 | No. 4 | April 2008
chromosome 7 polysomy in <43% of tumor cells responded to the drug, leading the authors to conclude that low EGFR gene copy number or low chromosome 7 polysomy have reduced likelihood to respond to panitumumab. Importantly, evaluation of patients treated with BSC alone showed no association between EGFR gene copy number or chromosome polysomy and progression-free survival [21]. The cut-off identified in our study may be equivalent to the cut-off proposed by Sartore-Bianchi et al. [21], since our analysis was conducted in 4 l tissue sections, while in the Sartore-Bianchi study the thickness was 2 l and higher copy numbers are expected in sections that include higher fraction of the nuclei [30]. Enumeration of centromere and/or gene copies in sections cut at widths lower than the diameter of intact nuclei is not exact since it only includes part of the nucleus. Therefore, the thinner tumor sections used in the Sartore-Bianchi study could be responsible for the lower cut-off best associated with drug sensitivity reported in their study. Differences in structure and mechanism of action between cetuximab and panitumumab could also account for the variation observed between the two studies. Although cetuximab and panitumumab target the same receptor, these two drugs present a different structure, since cetuximab is a chimeric immunoglobulin (Ig) G1 antibody and panitumumab is a fully human IgG2 antibody. Moreover, no studies have directly compared the two drugs and whether these agents are equally effective remain undefined. In addition to the studies mentioned in the previous paragraph, other studies recently explored the role of EGFR gene copy number in CRC treated with cetuximab [19, 31]. The largest investigation, conducted in 110 CRC patients, detected increased EGFR gene copy number at a frequency of 6% and found no association with disease control rate [31]. These negative results may be associated with the methodology used to assess EGFR gene copy number, quantitative PCR, which was deemed not optimal in lung cancer treated with EGFR-TKIs [32, 33]. Comparative analyses of biomarker status in primary tumor and corresponding metastasis have not been extensively carried
doi:10.1093/annonc/mdm492 | 721
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on April 30, 2012
Figure 2. Time to progression and survival in epidermal growth factor receptor fluorescence in situ hybridization (EGFR FISH) positive and negative, according to the cut-off of 2.92 identified with the receiver operating characteristic analysis. EGFR FISH-positive patients had a significantly longer time to progression (6.6 versus 3.5 months, P = 0.02) and a nonsignificant longer survival (11.3 versus 8.5 months, P = 0.8).
original article
funding Umbria Association Against Cancer to G.F.
acknowledgements We are indebted to the Tissue Procurement and Cytogenetics Cores of the University of Colorado Cancer Center for technical assistance with the immunohistochemistry and fluorescence in situ hybridization (FISH) assays and to Russell Nash for pathology assistance during the FISH analysis.
references 1. Jemal A, Siegel R, Ward E et al. Cancer statistics, 2007. CA Cancer J Clin 2007; 57: 43–66. 2. Thatcher N, Chang A, Parikh P et al. Gefitinib plus best supportive care in previously treated patients with refractory advanced non-small-cell lung cancer: results from a randomised, placebo-controlled, multicentre study (Iressa Survival Evaluation in Lung Cancer). Lancet 2005; 366: 1527–1537. 3. Shepherd FA, Rodrigues Pereira J, Ciuleanu T et al. Erlotinib in previously treated non-small-cell lung cancer. N Engl J Med 2005; 353: 123–132. 4. Kindler HL, Friberg G, Skoog L et al. Phase I/II trial of gefitinib and oxaliplatin in patients with advanced colorectal cancer. Am J Clin Oncol 2005; 28: 340–344. 5. 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: 337–345. 6. 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: 1658–1664. 7. Chung KY, Shia J, Kemeny NE et al. Cetuximab shows activity in colorectal cancer patients with tumors that do not express the epidermal growth factor receptor by immunohistochemistry. J Clin Oncol 2005; 23: 1803–1810. 8. Paez JG, Janne PA, Lee JC et al. EGFR mutations in lung cancer: correlation with clinical response to gefitinib therapy. Science 2004; 304: 1497–1500. 9. Cappuzzo F, Hirsch FR, Rossi E et al. Epidermal growth factor receptor gene and protein and gefitinib sensitivity in non-small-cell lung cancer. J Natl Cancer Inst 2005; 97: 643–655.
722 | Cappuzzo et al.
10. Cappuzzo F, Varella-Garcia M, Shigematsu H et al. Increased HER2 gene copy number is associated with response to gefitinib therapy in epidermal growth factor receptor-positive non-small-cell lung cancer patients. J Clin Oncol 2005; 23: 5007–5018. 11. Pao W, Wang TY, Riely GJ et al. KRAS mutations and primary resistance of lung adenocarcinomas to gefitinib or erlotinib. PLoS Med 2005; 2: e17. 12. Massarelli E, Varella-Garcia M, Tang X et al. Kras mutation is an important predictor of resistance to therapy with epidermal growth factor receptor tyrosine kinase inhibitors in patients with non-small cell lung cancer. Clin Cancer Res 2007; 13: 2890–2896. 13. Barber TD, Vogelstein B, Kinzler KW et al. Somatic mutations of EGFR in colorectal cancers and glioblastomas. N Engl J Med 2004; 351: 2883. 14. Perez-Torres M, Guix M, Gonzalez A et al. Epidermal growth factor receptor (EGFR) antibody down-regulates mutant receptors and inhibits tumors expressing EGFR mutations. J Biol Chem 2006; 281: 40183–40192. 15. Mukohara T, Engelman JA, Hanna NH et al. Differential effects of gefitinib and cetuximab on non-small-cell lung cancers bearing epidermal growth factor receptor mutations. J Natl Cancer Inst 2005; 97: 1185–1194. 16. Benvenuti S, Sartore-Bianchi A, Di Nicolantonio F et al. Oncogenic activation of the RAS/RAF signaling pathway impairs the response of metastatic colorectal cancers to anti-epidermal growth factor receptor antibody therapies. Cancer Res 2007; 67: 2643–2648. 17. 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: 1166–1169. 18. Lievre 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: 3992–3995. 19. Lenz HJ, Van Cutsem E, Khambata-Ford S et al. Multicenter phase II and translational study of cetuximab in metastatic colorectal carcinoma refractory to irinotecan, oxaliplatin, and fluoropyrimidines. J Clin Oncol 2006; 24: 4914–4921. 20. Moroni M, Veronese S, Benvenuti S et al. Gene copy number for epidermal growth factor receptor (EGFR) and clinical response to antiEGFR treatment in colorectal cancer: a cohort study. Lancet Oncol 2005; 6: 279–286. 21. Sartore-Bianchi A, Moroni M, Veronese S et al. Epidermal growth factor receptor gene copy number and clinical outcome of metastatic colorectal cancer treated with panitumumab. J Clin Oncol 2007; 25: 3238–3245. 22. Therasse P, Arbuck SG, Eisenhauer EA et al. New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 2000; 92: 205–216. 23. Varella-Garcia M. Stratification of non-small cell lung cancer patients for therapy with epidermal growth factor receptor inhibitors: the EGFR fluorescence in situ hybridization assay. Diagn Pathol 2006; 1: 19. 24. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1985; 53: 457–481. 25. Agresti A. Categorical Data Analysis. New York, NY: John Wiley & Sons 1990. 26. Cai T, Moskowitz CS. Semi-parametric estimation of the binomial ROC curve for a continuous diagnostic test. Biostatistics 2004; 5: 573–586. 27. Slamon DJ, Leyland-Jones B, Shak S et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med 2001; 344: 783–792. 28. Bonner JA, Harari PM, Giralt J et al. Radiotherapy plus cetuximab for squamouscell carcinoma of the head and neck. N Engl J Med 2006; 354: 567–578. 29. Moore MJ, Goldstein D, Hamm J et al. Erlotinib plus gemcitabine compared with gemcitabine alone in patients with advanced pancreatic cancer: a phase III trial of the National Cancer Institute of Canada Clinical Trials Group. J Clin Oncol 2007; 25: 1960–1966. 30. Berezuk M, West J, Varella-Garcia M et al. Adjusting interphase FISH results in epithelial tissue sections to whole cell complement. Anal Quant Cytol Histol 2001; 23: 93–100. 31. Khambata-Ford S, Garrett CR, Meropol NJ et al. Expression of epiregulin and amphiregulin and K-ras mutation status predict disease control in metastatic colorectal cancer patients treated with cetuximab. J Clin Oncol 2007; 25: 3230–3237.
Volume 19 | No. 4 | April 2008
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on April 30, 2012
out in CRC. Three previous studies evaluated EGFR expression in primary tumor and corresponding metastasis, one with concordant levels of expression [34] and two with discordant results [35, 36], similarly to our findings. Several reasons could be advocated to explain the different results. Subjectivity of IHC interpretation may be a key factor, and variations in fixation procedures and antibodies could affect the sensitivity of these assays, precluding comparison of results from distinct laboratories. Conversely, our study showed a significant concordance between different disease sites using analyses at the genomic level, which supports assessment of the EGFR genomic status by FISH indifferently in primary tumor or in distant sites. In conclusion, increased EGFR gene copy number assessed by FISH identified the chemotherapy refractory CRC patients most suitable for treatment with the anti-EGFR antibody cetuximab. The proposed cut-off value to be used in clinical practice for stratification of patients for this therapeutic strategy should be validated and refined in standardized methodological conditions and submitted to prospective validation.
Annals of Oncology
Annals of Oncology
32. Dziadziuszko R, Witta SE, Cappuzzo F et al. Epidermal growth factor receptor messenger RNA expression, gene dosage, and gefitinib sensitivity in non-small cell lung cancer. Clin Cancer Res 2006; 12: 3078–3084. 33. Bell DW, Lynch TJ, Haserlat SM et al. Epidermal growth factor receptor mutations and gene amplification in non-small-cell lung cancer: molecular analysis of the IDEAL/INTACT gefitinib trials. J Clin Oncol 2005; 23: 8081–8092. 34. Italiano A, Vandenbos FB, Otto J et al. Comparison of the epidermal growth factor receptor gene and protein in primary non-small-cell-lung cancer and
original article metastatic sites: implications for treatment with EGFR-inhibitors. Ann Oncol 2006; 17: 981–985. 35. Scartozzi M, Bearzi I, Berardi R et al. Epidermal growth factor receptor (EGFR) status in primary colorectal tumors does not correlate with EGFR expression in related metastatic sites: implications for treatment with EGFR-targeted monoclonal antibodies. J Clin Oncol 2004; 22: 4772–4778. 36. McKay JA, Murray LJ, Curran S et al. Evaluation of the epidermal growth factor receptor (EGFR) in colorectal tumors and lymph node metastases. Eur J Cancer 2002; 38: 2258–2264.
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on April 30, 2012
Volume 19 | No. 4 | April 2008
doi:10.1093/annonc/mdm492 | 723
Annals of Oncology 19: 717–723, 2008 doi:10.1093/annonc/mdm492 Published online 31 December 2007
EGFR FISH assay predicts for response to cetuximab in chemotherapy refractory colorectal cancer patients F. Cappuzzo1*, G. Finocchiaro1,2, E. Rossi3, P.A. Ja¨nne4, C. Carnaghi5, C. Calandri6, K. Bencardino7, C. Ligorio1, F. Ciardiello7, T. Pressiani1, A. Destro1,8, M. Roncalli1,8, L. Crino9, W. A. Franklin2, A. Santoro1 & M. Varella-Garcia1 1
Background: Standardized conditions to distinguish subpopulations of colorectal cancer (CRC) patients more and less sensitive to cetuximab therapy remain undefined.
Materials and methods: We retrospectively analyzed epidermal growth factor receptor (EGFR) copy number by fluorescence in situ hybridization (FISH) in paraffin-embedded tumor blocks from 85 chemorefractory CRC patients treated with cetuximab. Results were analyzed according to different score systems previously reported in colorectal and lung cancers. The primary end point of the study was identification of the EGFR FISH score that best associates with response rate (RR). Results: Using receiver operating characteristic (ROC) analysis, the cut-off that best discriminated responders versus nonresponders to cetuximab was a mean of 2.92 EGFR gene copies per cell. This model showed sensitivity of 58.6% [95% confidence interval (CI) = 47.1–70.1) and specificity of 93.3% (95% CI = 80.6–100). EGFR FISH-positive patients (N = 43, 50.6%) had significantly higher RR (P = 0.0001) and significantly longer time to disease progression (P = 0.02) than EGFR FISH negative (N = 42, 49.4%). Other scoring systems resulted less accurate in discriminating patients with the highest likelihood of response to cetuximab therapy. Conclusions: CRC patients with high EGFR gene copy number have an increased likelihood to respond to cetuximab therapy. Prospective clinical trials with a careful standardization of assay conditions and pattern interpretation are urgently needed. Key words: cetuximab, colon cancer, EGFR, fluorescence in situ hybridization
introduction Colorectal cancer (CRC) is one of the most common human malignancies and one of the leading causes of cancer-related death in the Western world [1]. During the last years, several strategies interfering with the epidermal growth factor receptor (EGFR) have been evaluated in solid tumors, including CRC. Although EGFR tyrosine kinase inhibitors (TKIs) demonstrated to induce dramatic and durable responses in certain patients with advanced non-small-cell lung cancer (NSCLC), no activity has been observed in CRC [2–4]. Conversely, monoclonal antibodies (mAbs) against EGFR, such as cetuximab (C225, Erbitux; ImClone System Inc, New York, NY) and panitumumab showed activity in metastatic CRC when used alone or in combination with irinotecan [5, 6]. *Correspondence to: Dr F. Cappuzzo, Istituto Clinico Humanitas, via Manzoni 56, 20089-Rozzano, Italy. Tel: +39-02-82244097; Fax: +39-02-82244097; E-mail: [email protected]
Biological mechanisms underlying response or resistance to antibodies against EGFR are only partially known. Although at the present time regulatory agencies require selections of patients for such therapy based on EGFR expression by immunohistochemistry (IHC), responses were also noted in patients whose tumors had no EGFR immunostaining [7], confirming the pitfalls of IHC in assessing EGFR expression and its limits for patient selection. In NSCLC, activating EGFR gene mutations and EGFR gene increased copy number are considered the most critical factors associated with response to TKI therapy, while KRAS mutations are responsible for intrinsic resistance to TKI in some patients [8–12]. In CRC, mutations in the tyrosine kinase domain of the EGFR gene occur at very low frequency [13] and their role in cetuximab sensitivity seem to be irrelevant [14, 15], while RAS and RAF mutations are associated with resistance to therapy with antibodies against EGFR [16–18].
ª The Author 2007. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For Permissions, please email: [email protected]
original article
Received 10 August 2007; revised 18 September 2007; accepted 19 September 2007
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on April 30, 2012
Department of Medical Oncology, Istituto Clinico Humanitas IRCCS, Rozzano, Italy; 2Department of Medicine/Medical Oncology, University of Colorado Cancer Center, Aurora, CO, USA; 3CINECA-Interuniversity Consortium, Bologna, Italy; 4Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA, USA; 5 Department of Medical Oncology, Bellaria-Maggiore Hospital, Bologna; 6Department of Medical Oncology, San Matteo Hospital, Pavia; 7Department of Medical Oncology, Seconda Universita’ di Napoli, Napoli; 8Pathology Unit, Milan University, Istituto Clinico Humanitas IRCCS, Rozzano; 9Department of Medical Oncology, Ospedale Silvestrini, Perugia, Italy
original article
materials and methods patient selection This retrospective study was conducted in a cohort of 85 CRC patients treated with cetuximab-based therapy for metastatic disease from January 2002 to August 2006 in five Italian institutions: Istituto Clinico Humanitas, Rozzano; Ospedale Bellaria, Bologna; Policlinico S.Matteo, Pavia; Ospedale Silvestrini, Perugia; and Seconda Universita` di Napoli, Napoli. For initial treatment in their institutions, patients were selected by presence of EGFR expression assessed in the local hospitals, using a commercially available kit (DAKO, Carpinteria, CA), and considering as positive all tumors with at least 1% immunostaining in tumor cells. For this study, patients were selected based on two main criteria: presence of at least one measurable lesion and availability of tumor tissue. No other clinical or biological criterion was used for patient selection. Cetuximab was given to each patient at the initial dose of 400 mg/m2, followed by weekly infusion of 250 mg/m2. In all patients, disease assessment was carried out every 2 months, with a confirmatory evaluation no <4 weeks after the response assessment, according to the response evaluation criteria in solid tumors (RECIST) criteria [22]. The study was approved by the local Ethics Committee and was conducted in accordance with ethical principles stated in the most recent version of the Declaration of Helsinki.
tissue preparation, FISH and IHC IHC and FISH assays were carried out in sections cut at 4 lm from paraffin-embedded tissue blocks containing representative malignant cells
718 | Cappuzzo et al.
and obtained at time of diagnosis. All IHC and FISH analyses were carried out at the University of Colorado Cancer Center, USA. One slide was hematoxylin–eosin (H&E) stained to be used as reference to select dominant tumor foci and sequential sections were assayed for the selected markers. The EGFR FISH assay was carried out with the LSI EGFR SpectrumOrange/CEP 7 SpectrumGreen probe (Vysis/Abbott Molecular, Des plaines, IL) according to methods described elsewhere [9, 23]. Four physically distant tumor areas were selected guided by the H&E-stained slide and the EGFR and CEP7 signals were counted in at least 30 nuclei per tumor area at ·1000 magnification. Microscope analysis was carried out independently by two observers (GF, MVG) blinded to the patients’ clinical characteristics and in case of discordance a consensus was reached in further analyses. In cases of heterogeneity among the tumor areas, only the area with the worst score was considered. For patients with tissue from primary tumor and corresponding metastasis, the sample from primary lesion was used to represent the patient in the correlative analyses with clinical outcome. Evaluation of EGFR protein expression by IHC was repeated for all specimens using a mAb against the external domain of the EGFR (Zymed Laboratories, San Francisco, CA). This analysis was carried out blindly to patients’ clinical characteristics and FISH results (by WAF) and tumor was considered positive when staining was observed in at least 1% of cells.
statistical analyses The primary end point of the study was the identification of the EGFR FISH score that best associates with response rate (RR). Secondary end points were association with time to disease progression (TTP), overall survival (OS) and concordance of EGFR profile between primary tumors and corresponding metastases. TTP was calculated from the time of first cetuximab infusion to time of disease progression or last disease assessment. OS was calculated from the time of first cetuximab infusion to patient death or last contact. Differences in RR were compared by Fisher’s exact test or v2 test. TTP, OS and the 95% confidence intervals (CIs) were evaluated by survival analysis using Kaplan–Meier method [24]. TTP and OS for the groups with negative and positive biomarker were compared using the log-rank test. Concordance between biological data in primary tumor and corresponding metastasis was calculated by weighted k statistic [25]. Statistical significance was set at <0.05 for each analysis. Analysis of receiver operating characteristic (ROC) curve was carried out with the aim of determining a cut-off point for EGFR gene copy number as a continuous variable [26]. EGFR sensitivity and specificity were expressed in terms of percentage and the value for which sensitivity and sensibility were the highest has been chosen as the best cut-off point. In addition to the score identified in the present study by ROC analysis and indicated as score A, tumors were classified as FISH positive or negative based on classifications previously reported in CRC by Sartore-Bianchi et al. [21] (score B, EGFR FISH positive = mean of ‡2.47 gene signals per nucleus and score C, CEP 7 polysomy positive = ‡43% of cells with ‡3 copies of CEP7) and in lung carcinomas [9, 23] by our group (score D, EGFR FISH positive = ‡40% of cells with ‡4 copies of EGFR or gene amplification). All statistical analyses were carried out using SPSS version 11.5.1 (SPSS Italia srl, Bologna, Italy).
results patient characteristics FISH analysis was successful in tumor samples from 85 chemorefractory patients. As shown in Table 1, our cohort presented clinical characteristics and outcome similar to CRC patients included in a large registration trial with cetuximab [5]
Volume 19 | No. 4 | April 2008
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on April 30, 2012
Recent studies indicated that EGFR gene copy number could influence response to anti-EGFR mAbs cetuximab or panitumumab therapy in CRC [18–21], although different techniques and scoring systems were adopted in each study, making direct comparison and assessment of clinical applicability practically impossible. In the study conducted by Lievre et al. [18], an increased EGFR gene copy number assessed by chromogenic in situ hybridization was significantly associated with an objective tumor response to cetuximab. However, the low number of EGFR-positive patients (three cases) precluded any firm conclusion. In the study carried out by Lenz et al. [19], EGFR gene gain assessed by PCR was significantly associated with longer survival, raising the question whether EGFR genomic gain represents a favorable prognostic factor. In the retrospective analysis conducted by Moroni et al. [20] where EGFR genomic gain was assessed by fluorescence in situ hybridization (FISH), there was a significant association between high EGFR copy numbers and better response to anti-EGFR mAbs cetuximab or panitumumab. In this study, in which patients were selected by clinical outcome (10 responders, 21 nonresponders), there was only one responder with <3 copies of EGFR per cell and only one nonresponder with >2.1 copies of EGFR per cell. More recently, the same group analyzed retrospectively 92 CRC patients treated with panitumumab plus best supportive care (BSC) or BSC alone and concluded that tumors with <2.47 EGFR copies per cell or <43% cells with chromosome 7 polysomy (‡3 chromosome enumeration probe 7 copies per cell) had a reduced likelihood of response to the therapy [21]. Because there is an urgent need for defining a method for selection of patients for anti-EGFR therapies, we investigated the association between EGFR gene copy numbers evaluated by FISH with response to cetuximab therapy in advanced CRC.
Annals of Oncology
original article
Annals of Oncology
Table 1. Patient characteristics Patient characteristics
Table 2. Biological characteristics Number
100 63.5 36.5
65.9 30.6 3.5 76.5 23.5 74.1 52.9 21.2 18.8 25.9 28.2 45.9 21.2 40 35.3 83.5 84.7 90.6 3.5 5.9
and all patients had received at least one chemotherapy line for metastatic disease. In the whole population, overall response rate was 17.7%, including one complete (1.2%) and 14 partial responses (16.5%). Median time to progression (TTP) was 5.4 months and median OS was 10.5 months. In the univariate analysis, no clinical characteristic was associated with response to cetuximab therapy. EGFR analysis was carried out in tissue from primary tumor only in 43 cases (50.6%), from metastasis only in 20 cases (23.5%) and from both primary tumor and corresponding metastasis in 22 cases (25.9%). Primary tumor tissues originated from colon in 48 cases and from rectum in 17 cases, while metastatic tissues were obtained from liver (25 cases), lung (eight cases) or other sites (nine cases).
EGFR analyses Patients were evaluated for EGFR protein expression by IHC and gene copy number by FISH (Table 2). EGFR IHC analysis was successfully carried out in 82 patients. Although all patients included in the present study had received cetuximab-based therapy, because were considered EGFR IHC positive, when tested in the local hospital using standard DAKO immunostaining, the analysis carried out in our central laboratory detected 16 patients lacking EGFR expression
Volume 19 | No. 4 | April 2008
Marker
Positive result (N/%)
EGFR IHC 66/80.5 (N = 82) EGFR FISH (N = 85) Score A 43/50.6 Score B 59/69.4 Score C 41/48.2 Score D 34/40.0
Negative result (N/%)
Reference
16/19.5
42/49.4 26/30.6 44/51.8 51/60.0
Present article Sartore-Bianchi et al. [21] Sartore-Bianchi et al. [21] Cappuzzo et al. [9], Varella-Garcia [23]
In the present study, using ROC analysis, a mean of 2.92 EGFR gene copy number qualified the tumor as FISH positive (score A). As defined by Sartori-Bianchi [21] in CRC, a mean of 2.47 EGFR gene copy number (score B) or chromosome 7 polysomy in >43% of cells (score C) qualified the tumor as FISH positive. According to the score system proposed in NSCLC, a tumor was defined as FISH positive when ‡40% of cells had ‡4 copies of EGFR or in presence of gene amplification (score D, [9, 23]). EGFR, epidermal growth factor receptor; IHC, immunohistochemistry; FISH, fluorescence in situ hybridization; ROC, receiver operating characteristic; CRC, colorectal cancer; NSCLC, non-small-cell lung cancer.
(EGFR IHC negative, Table 2). EGFR IHC-positive status was not associated with any clinical characteristic or end point (Table 3). Among the 85 patients, two (2.4%) had EGFR gene amplification in the tumor cells, while all others had balanced EGFR and CEP7 copy numbers. Thirty-four tumors had high level of genomic gain (mean EGFR = 4.2, ‡4 copies in ‡40% of nuclei) and 21 tumors had no or very low genomic gain for both tested targets (mean EGFR = 2.2). ROC analysis identified the mean of 2.92 EGFR copies per cell as the best cut-off value associated with patient outcome (Figure 1). Using this cut-off, 43 patients (50.6%) were classified as EGFR FISH positive. RR (32.6% versus 2.4%, P = 0.0001), TTP (6.6 versus 3.5 months, P = 0.02) and OS (11.3 versus 8.5 months, P = 0.8) favored EGFR FISH-positive individuals (Figure 2, Table 3). This model showed a sensitivity of 58.6% (95% CI = 47.1–70.1) and a specificity of 93.3% (95% CI = 80.6–100). Further, as presented in Table 3, our patient population was analyzed according to other scoring systems previously reported in colorectal carcinomas (scores B and C, [21]), and in lung carcinomas (score D, [9, 23]). Using score B [21], 59 patients (69.4%) were FISH positive. EGFR FISH positive had a significantly higher RR (23.7% versus 3.8%, P = 0.03), a trend towards longer TTP (6.1 versus 3.4 months, P = 0.06), with no difference in survival (9.8 versus 10.5 months, P = 0.6). Score B showed a 35.7% sensitivity (95% CI = 24.5–46.9) and 93.3% specificity (95% CI = 80.6–100). Using score C [21], 41 patients (48.2%) had chromosome 7 polysomy and were considered as FISH positive. No difference in RR (15.9% versus 19.5%), TTP (4.1 versus 6.1 months, P = 0.7) and survival (10.8 versus 9.5 months, P = 0.3) was observed between FISH negative and FISH positive, respectively. For this model, sensitivity was 52.9% (95% CI = 41.2–64.6) and specificity was
doi:10.1093/annonc/mdm492 | 719
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on April 30, 2012
Total evaluated 85 Gender Male 54 Female 31 Median age (year, range) 63.2 (29–79) Eastern Cooperative Oncology Group performance status 0 56 1 26 2 3 Primary tumor site Colon 65 Rectum 20 Metastatic sites Liver 63 Lung 45 Peritoneum 18 Other 16 Number of metastatic sites 1 22 2 24 ‡3 39 Previous chemotherapy lines 1 18 2 34 ‡3 30 Previous irinotecan 71 Previous oxaliplatin 72 Cetuximab + irinotecan 77 Cetuximab + oxaliplatin 3 Cetuximab alone 5
%
original article
Annals of Oncology
Table 3. Clinical outcome according to biological characteristics Patients (N/%)
Overall response (N/%)
Time to progression (months)
Overall survival (months)
Sensitivity (CI 95%)
Specificity (CI 95%)
EGFR IHC+ EGFR IHC2 P value EGFR FISH+ (score A) EGFR FISH– (score A) P value EGFR FISH+ (score B) EGFR FISH– (score B) P value CEP7 FISH+ (score C) CEP7 FISH– (score C) P value EGFR FISH+ (score D) EGFR FISH– (score D) P value
66/80.5 16/19.5
11/16.7 3/18.8 1.0 14/32.6 1/2.4 0.0001 14/23.7 1/3.8 0.03 8/19.5 7/15.9 0.6 8/23.5 7/13.7 0.2
4.6 6.6 0.2 6.6 3.5 0.02 6.1 3.4 0.06 6.1 4.1 0.7 6.5 4.1 0.3
9.2 13.7 0.3 11.3 8.5 0.8 9.8 10.5 0.6 9.5 10.8 0.3 9.5 10.8 0.2
19.1 (9.8–28.4)
78.6 (57.1–100)
58.6 (47.1–70.1)
93.3 (80.6–100)
35.7 (24.5–46.7)
93.3 (80.6–100)
52.9 (41.2–64.6)
53.3 (28.1–78.5)
62.9 (51.6–74.2)
53.3 (28.1–78.5)
43/50.6 42/49.4 59/69.4 26/30.6 41/48.2 44/51.8 34/40.0 51/60.0
Score A: ROC analysis; Scores B and C: Sartore-Bianchi et al. [21]; Score D: Cappuzzo et al. [9], Varella-Garcia [23]. CI, confidence interval; EGFR, epidermal growth factor receptor; IHC, immunohistochemistry; FISH, fluorescence in situ hybridization; ROC, receiver operating characteristic.
1,0
concordance in EGFR expression (IHC) was observed between primary tumor and corresponding metastasis (P = 0.08). As shown in Table 4, among the 17 primary tumors positive for EGFR by IHC, 16 were also positive in corresponding metastasis and one was negative. Among the five primary tumors with negative EGFR IHC score, four resulted positive in the corresponding metastasis. The EGFR gene copy results by FISH were similar in primary tumor and metastasis, with a significant concordance independently of the score system adopted (P = 0.001, Table 4). Using the cut-off identified by the ROC analysis, 21 patients had the same EGFR gene score, and only in one case the tumor was classified as FISH positive in the primary lesion and FISH negative in the corresponding liver metastasis.
AUC=0.76
Sensitivity
,8
,5
,3
0,0 0,0
,3
,5
,8
1,0
Specificity Figure 1. Receiver operating characteristic analysis based on mean EGFR gene copy number with response to cetuximab therapy as end point. In this model, sensitivity was 58.6% and specificity was 93.3%.
53.3% (28.1–78.5). Using score D [9, 23], 34 patients (40.0%) were classified as EGFR FISH positive and no association was detected between FISH positive and clinical characteristic or outcome such as RR (P = 0.2), TTP (p = 0.3) or OS (P = 0.2).
biomarker results in primitive tumor and corresponding metastasis In order to investigate whether EGFR status differed from primary tumor and corresponding metastases, we independently assessed paired samples from 22 patients. No
720 | Cappuzzo et al.
discussion Anti-EGFR strategies are offering new hope for patients with cancer, as demonstrated by the survival improvement obtained in patients with lung [3], breast [27], head and neck [28], pancreatic [29] and CRC [5]. However, identification of molecular predictors of anti-EGFR therapies remains a challenge. Predictive factors for sensitivity or resistance to EGFR-TKIs have been identified in lung cancer, but scarce and conflicting data exist in other diseases, including gastrointestinal cancers. In the present translational study, one of the largest so far conducted in CRC exposed to cetuximab, we showed that EGFR genomic gain was associated with response to the drug, and a mean of 2.92 EGFR gene copy number best discriminated the sensitive population. Using this cut-off value, the positive predictive value was 97.6% and the negative predictive value was 32.5%, with about 3-month gain in TTP and survival, suggesting that this patient stratum is particularly suitable for anti-EGFR therapy.
Volume 19 | No. 4 | April 2008
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on April 30, 2012
Marker
original article 1,0
1,0 ,8
Cumulative Survival Function
Cumulative Distribution Function
Annals of Oncology
p=0.02
,6 EGFR FISH+
,4
,2 0,0
EGFR FISH-
0
10
20
Time to Progression (months)
,8
p=0.8
,6 EGFR FISH+
,4 EGFR FISH-
,2 0,0
0
10
20
30
Survival (months)
Table 4. Biomarker results in primary tumor and corresponding metastasis from 22 patients Marker
Metastasis
EGFR IHC+ EGFR IHC2 EGFR FISH+ (Score A) EGFR FISH2 (Score A) EGFR FISH+ (Score B) EGFR FISH2 (Score B) EGFR FISH+ (Score C) EGFR FISH2 (Score C) EGFR FISH+ (Score D) EGFR FISH2 (Score D)
Positive 16 4 12 0 14 1 14 1 10 1
Primary
Negative 1 1 1 9 0 7 1 6 2 9
K P value 0.08 0.001 0.001 0.001 0.001
Score A: ROC analysis; Scores B and C: Sartore-Bianchi et al. [21]; Score D: Cappuzzo et al. [9], Varella-Garcia [23]. EGFR, epidermal growth factor receptor; IHC, immunohistochemistry; FISH, fluorescence in situ hybridization; ROC, receiver operating characteristic.
Present results are confirmatory of previous studies in CRC that showed that EGFR genomic gain assessed by FISH was associated with response to the mAbs cetuximab or panitumumab [20, 21]. In the study carried out by Moroni et al. where 31 patients with metastatic CRC treated with cetuximab or panitumumab were evaluated for EGFR by FISH, increased EGFR gene copy number was observed in eight of nine patients with objective response and in only one of 21 nonresponders. Nevertheless, no cut-off for discriminating EGFR FISH positive versus negative was proposed precluding attempts to reproduce these results [20]. More recently, in order to define the best cut-off discriminating responders versus nonresponders, the same group analyzed by FISH a cohort of 92 CRC patients, including 58 cases treated with panitumumab and 34 treated with BSC alone [21]. In that study, a cut-off of 2.47 mean EGFR gene copy number best discriminated nonresponders, and none of the patients with
Volume 19 | No. 4 | April 2008
chromosome 7 polysomy in <43% of tumor cells responded to the drug, leading the authors to conclude that low EGFR gene copy number or low chromosome 7 polysomy have reduced likelihood to respond to panitumumab. Importantly, evaluation of patients treated with BSC alone showed no association between EGFR gene copy number or chromosome polysomy and progression-free survival [21]. The cut-off identified in our study may be equivalent to the cut-off proposed by Sartore-Bianchi et al. [21], since our analysis was conducted in 4 l tissue sections, while in the Sartore-Bianchi study the thickness was 2 l and higher copy numbers are expected in sections that include higher fraction of the nuclei [30]. Enumeration of centromere and/or gene copies in sections cut at widths lower than the diameter of intact nuclei is not exact since it only includes part of the nucleus. Therefore, the thinner tumor sections used in the Sartore-Bianchi study could be responsible for the lower cut-off best associated with drug sensitivity reported in their study. Differences in structure and mechanism of action between cetuximab and panitumumab could also account for the variation observed between the two studies. Although cetuximab and panitumumab target the same receptor, these two drugs present a different structure, since cetuximab is a chimeric immunoglobulin (Ig) G1 antibody and panitumumab is a fully human IgG2 antibody. Moreover, no studies have directly compared the two drugs and whether these agents are equally effective remain undefined. In addition to the studies mentioned in the previous paragraph, other studies recently explored the role of EGFR gene copy number in CRC treated with cetuximab [19, 31]. The largest investigation, conducted in 110 CRC patients, detected increased EGFR gene copy number at a frequency of 6% and found no association with disease control rate [31]. These negative results may be associated with the methodology used to assess EGFR gene copy number, quantitative PCR, which was deemed not optimal in lung cancer treated with EGFR-TKIs [32, 33]. Comparative analyses of biomarker status in primary tumor and corresponding metastasis have not been extensively carried
doi:10.1093/annonc/mdm492 | 721
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on April 30, 2012
Figure 2. Time to progression and survival in epidermal growth factor receptor fluorescence in situ hybridization (EGFR FISH) positive and negative, according to the cut-off of 2.92 identified with the receiver operating characteristic analysis. EGFR FISH-positive patients had a significantly longer time to progression (6.6 versus 3.5 months, P = 0.02) and a nonsignificant longer survival (11.3 versus 8.5 months, P = 0.8).
original article
funding Umbria Association Against Cancer to G.F.
acknowledgements We are indebted to the Tissue Procurement and Cytogenetics Cores of the University of Colorado Cancer Center for technical assistance with the immunohistochemistry and fluorescence in situ hybridization (FISH) assays and to Russell Nash for pathology assistance during the FISH analysis.
references 1. Jemal A, Siegel R, Ward E et al. Cancer statistics, 2007. CA Cancer J Clin 2007; 57: 43–66. 2. Thatcher N, Chang A, Parikh P et al. Gefitinib plus best supportive care in previously treated patients with refractory advanced non-small-cell lung cancer: results from a randomised, placebo-controlled, multicentre study (Iressa Survival Evaluation in Lung Cancer). Lancet 2005; 366: 1527–1537. 3. Shepherd FA, Rodrigues Pereira J, Ciuleanu T et al. Erlotinib in previously treated non-small-cell lung cancer. N Engl J Med 2005; 353: 123–132. 4. Kindler HL, Friberg G, Skoog L et al. Phase I/II trial of gefitinib and oxaliplatin in patients with advanced colorectal cancer. Am J Clin Oncol 2005; 28: 340–344. 5. 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: 337–345. 6. 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: 1658–1664. 7. Chung KY, Shia J, Kemeny NE et al. Cetuximab shows activity in colorectal cancer patients with tumors that do not express the epidermal growth factor receptor by immunohistochemistry. J Clin Oncol 2005; 23: 1803–1810. 8. Paez JG, Janne PA, Lee JC et al. EGFR mutations in lung cancer: correlation with clinical response to gefitinib therapy. Science 2004; 304: 1497–1500. 9. Cappuzzo F, Hirsch FR, Rossi E et al. Epidermal growth factor receptor gene and protein and gefitinib sensitivity in non-small-cell lung cancer. J Natl Cancer Inst 2005; 97: 643–655.
722 | Cappuzzo et al.
10. Cappuzzo F, Varella-Garcia M, Shigematsu H et al. Increased HER2 gene copy number is associated with response to gefitinib therapy in epidermal growth factor receptor-positive non-small-cell lung cancer patients. J Clin Oncol 2005; 23: 5007–5018. 11. Pao W, Wang TY, Riely GJ et al. KRAS mutations and primary resistance of lung adenocarcinomas to gefitinib or erlotinib. PLoS Med 2005; 2: e17. 12. Massarelli E, Varella-Garcia M, Tang X et al. Kras mutation is an important predictor of resistance to therapy with epidermal growth factor receptor tyrosine kinase inhibitors in patients with non-small cell lung cancer. Clin Cancer Res 2007; 13: 2890–2896. 13. Barber TD, Vogelstein B, Kinzler KW et al. Somatic mutations of EGFR in colorectal cancers and glioblastomas. N Engl J Med 2004; 351: 2883. 14. Perez-Torres M, Guix M, Gonzalez A et al. Epidermal growth factor receptor (EGFR) antibody down-regulates mutant receptors and inhibits tumors expressing EGFR mutations. J Biol Chem 2006; 281: 40183–40192. 15. Mukohara T, Engelman JA, Hanna NH et al. Differential effects of gefitinib and cetuximab on non-small-cell lung cancers bearing epidermal growth factor receptor mutations. J Natl Cancer Inst 2005; 97: 1185–1194. 16. Benvenuti S, Sartore-Bianchi A, Di Nicolantonio F et al. Oncogenic activation of the RAS/RAF signaling pathway impairs the response of metastatic colorectal cancers to anti-epidermal growth factor receptor antibody therapies. Cancer Res 2007; 67: 2643–2648. 17. 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: 1166–1169. 18. Lievre 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: 3992–3995. 19. Lenz HJ, Van Cutsem E, Khambata-Ford S et al. Multicenter phase II and translational study of cetuximab in metastatic colorectal carcinoma refractory to irinotecan, oxaliplatin, and fluoropyrimidines. J Clin Oncol 2006; 24: 4914–4921. 20. Moroni M, Veronese S, Benvenuti S et al. Gene copy number for epidermal growth factor receptor (EGFR) and clinical response to antiEGFR treatment in colorectal cancer: a cohort study. Lancet Oncol 2005; 6: 279–286. 21. Sartore-Bianchi A, Moroni M, Veronese S et al. Epidermal growth factor receptor gene copy number and clinical outcome of metastatic colorectal cancer treated with panitumumab. J Clin Oncol 2007; 25: 3238–3245. 22. Therasse P, Arbuck SG, Eisenhauer EA et al. New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 2000; 92: 205–216. 23. Varella-Garcia M. Stratification of non-small cell lung cancer patients for therapy with epidermal growth factor receptor inhibitors: the EGFR fluorescence in situ hybridization assay. Diagn Pathol 2006; 1: 19. 24. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1985; 53: 457–481. 25. Agresti A. Categorical Data Analysis. New York, NY: John Wiley & Sons 1990. 26. Cai T, Moskowitz CS. Semi-parametric estimation of the binomial ROC curve for a continuous diagnostic test. Biostatistics 2004; 5: 573–586. 27. Slamon DJ, Leyland-Jones B, Shak S et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med 2001; 344: 783–792. 28. Bonner JA, Harari PM, Giralt J et al. Radiotherapy plus cetuximab for squamouscell carcinoma of the head and neck. N Engl J Med 2006; 354: 567–578. 29. Moore MJ, Goldstein D, Hamm J et al. Erlotinib plus gemcitabine compared with gemcitabine alone in patients with advanced pancreatic cancer: a phase III trial of the National Cancer Institute of Canada Clinical Trials Group. J Clin Oncol 2007; 25: 1960–1966. 30. Berezuk M, West J, Varella-Garcia M et al. Adjusting interphase FISH results in epithelial tissue sections to whole cell complement. Anal Quant Cytol Histol 2001; 23: 93–100. 31. Khambata-Ford S, Garrett CR, Meropol NJ et al. Expression of epiregulin and amphiregulin and K-ras mutation status predict disease control in metastatic colorectal cancer patients treated with cetuximab. J Clin Oncol 2007; 25: 3230–3237.
Volume 19 | No. 4 | April 2008
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on April 30, 2012
out in CRC. Three previous studies evaluated EGFR expression in primary tumor and corresponding metastasis, one with concordant levels of expression [34] and two with discordant results [35, 36], similarly to our findings. Several reasons could be advocated to explain the different results. Subjectivity of IHC interpretation may be a key factor, and variations in fixation procedures and antibodies could affect the sensitivity of these assays, precluding comparison of results from distinct laboratories. Conversely, our study showed a significant concordance between different disease sites using analyses at the genomic level, which supports assessment of the EGFR genomic status by FISH indifferently in primary tumor or in distant sites. In conclusion, increased EGFR gene copy number assessed by FISH identified the chemotherapy refractory CRC patients most suitable for treatment with the anti-EGFR antibody cetuximab. The proposed cut-off value to be used in clinical practice for stratification of patients for this therapeutic strategy should be validated and refined in standardized methodological conditions and submitted to prospective validation.
Annals of Oncology
Annals of Oncology
32. Dziadziuszko R, Witta SE, Cappuzzo F et al. Epidermal growth factor receptor messenger RNA expression, gene dosage, and gefitinib sensitivity in non-small cell lung cancer. Clin Cancer Res 2006; 12: 3078–3084. 33. Bell DW, Lynch TJ, Haserlat SM et al. Epidermal growth factor receptor mutations and gene amplification in non-small-cell lung cancer: molecular analysis of the IDEAL/INTACT gefitinib trials. J Clin Oncol 2005; 23: 8081–8092. 34. Italiano A, Vandenbos FB, Otto J et al. Comparison of the epidermal growth factor receptor gene and protein in primary non-small-cell-lung cancer and
original article metastatic sites: implications for treatment with EGFR-inhibitors. Ann Oncol 2006; 17: 981–985. 35. Scartozzi M, Bearzi I, Berardi R et al. Epidermal growth factor receptor (EGFR) status in primary colorectal tumors does not correlate with EGFR expression in related metastatic sites: implications for treatment with EGFR-targeted monoclonal antibodies. J Clin Oncol 2004; 22: 4772–4778. 36. McKay JA, Murray LJ, Curran S et al. Evaluation of the epidermal growth factor receptor (EGFR) in colorectal tumors and lymph node metastases. Eur J Cancer 2002; 38: 2258–2264.
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on April 30, 2012
Volume 19 | No. 4 | April 2008
doi:10.1093/annonc/mdm492 | 723