433
Cancer Biomarkers 15 (2015) 433–440 DOI 10.3233/CBM-150483 IOS Press
MMP-9/ANC score as a predictive biomarker for efficacy of bevacizumab plus platinum doublet chemotherapy in patients with advanced or recurrent non-squamous non-small cell lung cancer Kazuya Hiuraa,b , Akiko Shiraishia , Chinami Suzukic, Kei Takamurad , Makoto Yamamotod , Hitoshi Komorie , Yasuhiro Watanabea and Sachiko Iwaki-Egawaa,∗ a
Division of Life Science, Hokkaido Pharmaceutical University School of Pharmacy, Sapporo, Japan Department of Hospital Pharmacy, Abashiri Kosei General Hospital, Abashiri, Japan c Department of Hospital Pharmacy, Engaru Kosei General Hospital, Engaru, Japan d First Department of Internal Medicine, Obihiro Kosei General Hospital, Obihiro, Japan e Department of Hospital Pharmacy, Obihiro Kosei General Hospital, Obihiro, Japan b
Abstract. BACKGROUND: Bevacizumab is a recombinant humanized monoclonal antibody against vascular endothelial growth factor (VEGF), which is a key regulator of tumor angiogenesis. OBJECTIVE: To evaluate biomarkers to predict the benefit of paclitaxel and carboplatin plus bevacizumab (PCB) therapy in patients with advanced or recurrent non-squamous non-small cell lung cancer. METHODS: Among 21 patients treated with PCB, 10 were included in the good responder group and 11 in the non-responder group. Serum VEGF, MMP-2 and MMP-9 were measured using ELISA. RESULTS: There were no significant differences in these markers levels between groups. However, the good responder group showed a significantly higher pre-treatment MMP-9/ absolute neutrophil count (ANC) score than the non-responder group before the treatment (p = 0.014), and there was a positive correlation between the score and the tumor reduction rate (r = 0.57, p = 0.016). Furthermore, by dividing patients into a high scoring group (MMP-9/ANC median, n = 11) and a low scoring group (MMP-9/ANC < median, n = 10), former group showed a significant improvement in the median progression-free survival compared with latter group (636 vs. 196 days, p = 0.032). CONCLUSIONS: MMP-9/ANC score before PCB treatment may be a suitable biomarker to assess the anti-tumor effects of PCB therapy. Keywords: Absolute neutrophil count, matrix metalloproteinase-9, progression-free survival, tumor reduction rate, VEGF
1. Introduction ∗ Corresponding author: Sachiko Iwaki-Egawa, Division of Life Science, Hokkaido Pharmaceutical University, School of Pharmacy, 7-15-4-1 Maeda, Teine, Sapporo 006-8590, Hokkaido, Japan. Tel.: +81 11 676 8700; Fax: +81 11 676 8666; E-mail: sachiko@ hokuyakudai.ac.jp.
Bevacizumab (BEV) is a recombinant humanized monoclonal antibody against vascular endothelial growth factor (VEGF), which is a key regulator of tumor angiogenesis. BEV promotes its anti-angiogenic
c 2015 – IOS Press and the authors. All rights reserved ISSN 1574-0153/15/$35.00
434
K. Hiura et al. / MMP-9/ANC score as a predictive biomarker for efficacy of bevacizumab plus platinum doublet chemotherapy
effect by preventing soluble VEGF from binding to its specific receptors, thereby blocking and neutralizing VEGF-induced biological events [1,2]. As this mechanism of action is different from that of standard chemotherapeutic agents, BEV is not expected to cause typical chemotherapy-related toxicity or exacerbate the toxicity of concomitant chemotherapeutic agents. However, several principal BEV-related adverse events, such as arterial hypertension, proteinuria, and intestinal hemorrhage, have been observed in clinical trials; thus, oncologists must pay close attention to signs of these side effects [3]. Some recent reports have declared that BEV-induced hypertension may be a predictive marker for anti-angiogenic treatment efficacy [4–6]. A previous study reported that the occurrence of hypertension was associated with longer duration of the progression-free survival (PFS) and good overall survival (OS) compared with non-hypertensive patients with non-small cell lung cancer (NSCLC) receiving chemotherapy with paclitaxel and carboplatin plus BEV (PCB) [6]. There are, however, some problems with these findings. For example, it remains unclear whether a history of hypertension management can affect the predictive value of BEV treatment efficacy or whether the compensatory mechanisms to change blood pressure or multiple definitions of hypertension used in the study affected the results [7]. Therefore, we cannot definitively conclude that the onset of hypertension during PCB treatment is a clear predictor of improved outcome. VEGF binds to its receptors that are expressed on endothelial cell membranes, and produces and activates proteases, including matrix metalloproteinases (MMPs) [8], which increase vascular permeability by degrading the basement membrane and extracellular matrix (ECM), and promotes endothelial cells migration to the ECM and subsequent proliferation to induce new vascular grow. MMP-2 (72 kDa) and MMP-9 (92 kDa) are two MMP subtypes that are known to be involved in the degradation of type IV collagen, which composes the basement membrane of the capillary endothelium [9–11]. In the present study, we assessed the suitability of VEGF, MMP-2, MMP-9, and MMP-9/absolute neutrophil count (ANC) score as candidate biomarkers to predict the benefit of PCB chemotherapy in patients with advanced or recurrent non-squamous NSCLC.
2. Patients and methods 2.1. Patients Twenty-one patients in advanced or recurrent NSCLS were enrolled and treated with BEV plus platinum doublet combination chemotherapy on a 21-day treatment cycle from August 2010 to October 2011. According to the Response Evaluation Criteria in Solid Tumors (RECIST) guidelines [12], after the treatment patients were classified as patients with a complete response (CR), partial response (PR), stable disease (SD) or progressive disease (PD). A good responder group was deemed as the patients with CR or PR, and a nonresponder group was deemed as the patients with SD or PD. In this study, post-treatment means the time the evaluation was done. The primary endpoint of the study was response rate and tumor reduction rate. Secondary endpoints included PFS, OS, and toxicity. Concentrations of serum VEGF, MMP-2, and MMP-9 were measured at pre- and post-treatment with BEV and platinum doublet. As a control, healthy subjects were provided by the Hokkaido Red Cross Blood Center (Sapporo, Japan). Samples were stored at −30◦C or −80◦C until use. This study was performed in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of Obihiro Kosei General Hospital. Full, written, and informed consent was also obtained. 2.2. Enzyme-linked immunosorbent assay (ELISA) The following biomarkers were measured according to the manufacture’s instructions: VEGF (human VEGF Quantikine; R&D Systems, Minneapolis, MN, USA), MMP-2, and MMP-9 (human ELISA Biotrak; GE Healthcare, Buckinghamshire, UK). 2.3. Gelatin zymography Gelatinolytic activities of MMPs were analyzed using gelatin zymography [13]. Serum (equivalent to 0.5 µL) was run, without reduction, in a 0.1% SDS, 7.5% polyacrylamide gel impregnated with 0.3 mg/ml gelatin. Gel electrophoresis was performed at 4◦ C. After electrophoresis, the gel was washed in 2.5% Triton X-100 for 15 min and in 50 mM Tris-HCl, pH7.6 for 10 min. Then it was incubated overnight at 37◦ C in a buffer containing 30 mM Tris-HCl, pH 7.6, 5 mM CaCl2 , 0.2 M NaCl, and 0.02% NaN3 . Following this incubation, the gel was stained for proteins with 0.1%
K. Hiura et al. / MMP-9/ANC score as a predictive biomarker for efficacy of bevacizumab plus platinum doublet chemotherapy 3,000 2,500 VEGF (pg/mL)
Coomassie Blue R250 in 40% (v/v) isopropanol and destained in 7% (v/v) acetic acid. Gel images were acquired with a scanner, and MMP levels were quantified by densitometric analysis of the bands by Image J 1.47d (National Institutes of Health, USA).
435
2,000 1,500 1,000 500
2.4. Statistical analysis
0 CR + PR (n = 10)
CR + PR (n = 10)
SD + PD (n = 11)
P o st- t r e a t me n t
Fig. 1. Box-plots of serum VEGF levels in pre- and post-treatment with PCB chemotherapy. The ends of the whiskers represent the minimum and maximum of all of the data. Inbox bars show median for VEGF level of each group. Post-treatment means the time the evaluation was done. 10,000
3. Results
7,500
5,000
2,500
0 CR + PR (n = 10)
SD + PD (n = 11)
CR + PR (n = 10)
Pre -t reat me nt
SD + PD (n = 11)
Pos t-treatment
A 2,000
MMP-9 (ng/mL)
Of 21 patients who received BEV plus platinum doublet combination chemotherapy, 10 patients were classified as good responders (CR/PR, 1/9) and 11 as non-responders (SD/PD, 10/1). There were no significant differences between groups with regards to age, gender, performance status (PS), histology, clinical stage, mutations of the epidermal growth factor receptor gene (EGFR-mt), Brinkman index, or hypertension (i.e., high blood pressure by the end of cycle 1 was defined as blood pressure > 150/100-mmHg at any previous time or at least a 20-mmHg increase in diastolic blood pressure from baseline) (Table 1). There were no significant differences in serum VEGF levels between groups both pre- and posttreatment (Fig. 1). Pre-treatment VEGF levels were much higher in both groups than in healthy subjects (228 ± 144 pg/mL), and post-treatment, both decreased nearly to the levels of that observed in healthy subjects; however, there were no statistical differences between groups. Pre-treatment serum MMP-2 levels of both groups were almost the same as those of healthy subjects (2051 ± 228 ng/mL) and there were no significant differences in pre- and post-treatment serum MMP-2 levels between or within groups (Fig. 2A). Furthermore, there were no significant differences in serum MMP-9 levels, as with serum MMP-2 levels, between groups both pre- and post-treatment (Fig. 2B). How-
SD + PD (n = 11) P r e - tr e a t me n t
MMP-2 (ng/mL)
The data were analyzed with Student’s t-test, Welch’s test, Paired t-test in the parametric test, MannWhitney U test, Wilcoxon signed rank test in the nonparametric test, and Fisher’s exact probability test in the regression analysis. Spearman rank correlation coefficient was used for the analysis of the correlation. Survival probability was estimated using the Kaplan-Meier method, and the log-rank test was used to evaluate survival distributions among different patient groups. P values < 0.05 were considered statistically significant.
1,500
1,000
500
0 CR + PR (n = 10)
SD + PD (n = 11)
CR + PR (n = 10)
SD + PD (n = 11)
Post-treatment
Pre-treatment
B
Fig. 2. Box-plots of serum MMP-2 and MMP-9 levels in pre- and post-treatment with PCB chemotherapy. The ends of the whiskers represent the minimum and maximum of all of the data. Inbox bars show median for MMP-2 level (A) and MMP-9 level (B) of each group. Post-treatment means the time the evaluation was done.
ever, serum MMP-9 levels were significantly higher in both groups than in healthy subjects (11.7 ± 6.3 ng/mL), and the good responder group showed a tendency of higher pre-treatment MMP-9 levels than the non-responder group. Then, we compared serum MMP-2 and MMP-9 levels using gelatin zymography and ELISA. A typical
436
K. Hiura et al. / MMP-9/ANC score as a predictive biomarker for efficacy of bevacizumab plus platinum doublet chemotherapy Table 1 Patient characteristics Number of patients Age, range (median) Gender, males/females ECOG PS, 0/1/2 Histology, Ad/other Stage, IIIA/IIIB/IV EGFR-mt, +/−/unknown Brinkman index, range (median) Hypertension, +/−
CR + PR CR/PR (1/9) 46–77 (66) 5/5 4/4/2 8/2 0/2/8 2/8/0 0-1, 650 (80) 8/2
SD + PD SD/PD (10/1) 40–79 (57) 7/4 6/5/0 9/2 3/0/8 2/7/2 0-1, 200 (700) 7/4
P value 0.30a 0.42b 0.29c 0.67b 0.62c 0.67b 0.14a 0.37b
Abbreviations: CR = complete response; PR = partial response; SD = stable disease; PD = progressive disease; ECOG = Eastern Cooperative Oncology Group; PS = performance status; Ad = adenocarcinoma; EGFR-mt = epidermal growth factor receptor mutation. a Student’s t-test; b Fisher’s exact probability test; c Mann-Whitney’s U test.
zymographic pattern of serum collected from a patient who was classified as SD is shown in Fig. 3A. In zymography, only latent form bands of MMP-2 and MMP-9 (proMMP-2 and proMMP-9, respectively) were observed. The intensities of the MMP-9 bands decreased throughout the treatment course, though there was no significant change between pre- and posttreatment as we showed in Fig. 2B. The reason is that almost the evaluation was done at 2nd or 3rd treatment course and there was statistically no difference. In addition, the intensities of the MMP-2 bands remained virtually unchanged (Fig. 3A). Quantitation of the bands showed that MMP-9 before treatment had a 5.3-fold higher activity against gelatin than MMP-2. The serum MMP-2 and MMP-9 levels as determined by ELISA in this same patient throughout the treatment course are shown in Fig. 3B. The ELISA kits employed in this study recognize only the precursors of MMP2 and MMP-9. In addition, the ELISA results showed that MMP-9 levels were decreased and MMP-2 levels remained unchanged following treatment. The absolute amount of MMP-2 was, however, over 10-fold greater than that of MMP-9. The difference in intensities between zymography and ELISA may exist in the specific activity of both MMP-2 and MMP-9 against the gelatin. The results of other patient serum samples were almost the same (data not shown). Serum MMP-9 levels were corrected by ANC (Fig. 4A) because MMP-9 is produced by blood cells, such as neutrophils [14]. The good responder group showed a significantly higher pre-treatment MMP9/ANC score than the non-respoder group (p = 0.014), indicating that the treatment efficacy will be superior if the basal MMP-9/ANC score is high. There was a positive correlation (r = 0.57, p = 0.016) between the pre-treatment MMP-9/ANC scores and the tumor re-
duction rates of 19 patients (CR + PR, n = 10; SD + PD, n = 9) with primary tumor that could be measured after treatment (Fig. 4B). We divided the patients into a high scoring group (MMP-9/ANC median, n = 11) and a low scoring group (MMP-9/ANC < median, n = 10) and evaluated PFS and OS in each. The high MMP-9/ANC scoring group demonstrated a significant improvement in median PFS compared with the low MMP-9/ANC scoring group (636 vs. 196 days, p = 0.032) (Fig. 5A), but there was no improvement in median OS (p = 0.558) (Fig. 5B).
4. Discussion Several biomarkers have been reported as predictors of responsiveness to BEV therapy including plasma VEGF-A levels and tumor expression of neuropilin1 in gastric cancer [15], as well as the number of peripheral endothelial cells in colorectal cancer [16]. At present, however, these biomarkers have not yet been fully elucidated; therefore, we examined several biomarkers to predict the curative effects of BEV in advanced or recurrent NSCLC. No significant differences were found in age or gender, PS, histology, clinical stage, mutations of the EGFR-mt, or Brinkman index between the good responder and non-responder groups. There was also no significant difference between the two groups in the prevalence of early hypertension, which is reported to be a predictive marker of the curative effect of BEV treatment. Hurwitz et al. recently reported results in accordance with those of the present study [17]. However, some studies reported that the hypertension in response to VEGF inhibition was associated with a
K. Hiura et al. / MMP-9/ANC score as a predictive biomarker for efficacy of bevacizumab plus platinum doublet chemotherapy 1
2
3
4
5
437
P =0.014
6
proMMP-9
MMP-9/ANC score
200
150
100
50
proMMP-2 0 CR + PR (n = 10)
SD + PD (n = 11)
CR + PR (n = 10)
Pre-t re atm ent
SD + PD (n = 11) P ost-tr ea tm ent
A
A
100
300
5,000
4,000
200 3,000 150 2,000
MMP-2 ( , ng/mL)
MMP-9 ( , ng/mL)
250
Tumor reduction rate (%)
80
60
40
20
0
100
0
50
100
150
200
MMP-9/ANC score
1,000
50
0
0 Pretreatment 2nd
3rd
4th
5th
6th
Treatment course
B Fig. 3. MMP-2 and MMP-9 levels of serum collected from a patient who was classified as SD. (A) A typical zymographic pattern of serum. Lanes 1-6 contain an equivalent of 0.5 µL serum. Only latent form bands of MMP-2 and MMP-9 (proMMP-2, 72 kDa and proMMP-9, 92 kDa, respectively) were observed. Lane 1: pre-treatment, lane 2: before 2nd course treatment, lane 3: before 3rd course, lane 4: before 4th course, lane 5: before 5th course, and lane 6: before 6th course. (B) Concentrations of serum MMP-2 and MMP-9 of the same patient detected by ELISA. Open circle: MMP-9, closed circle: MMP-2.
significant reduction in nitric oxide (NO) production, which is caused by inhibition of NO synthase as a result of endothelial cell injury [18]. In our preliminary study, we also measured serum NO levels, but found no significant differences between groups pre- and posttreatment (data not shown). Jubb et al. reported that the level of VEGF expression was neither a prognostic nor a predictive marker of benefit from BEV treatment in colon cancer [19]. Hegde et al. also indicated that a
B
Fig. 4. MMP-9/ANC score in pre- and post-treatment and correlation between the tumor reduction rates and the pretreatment MMP-9/ANC scores. (A) Box-plots of MMP-9/ANC score in pre- and post-treatment with PCB chemotherapy. The ends of the whiskers represent the minimum and maximum of all of the data. Inbox bars show median for MMP-9/ANC score of each group. (B) Correlation between the tumor reduction rates and the pre-treatment MMP-9/ANC scores. A line represents the linear regression (r = 0.57, p < 0.05). Open circle: CR + PR (n = 10), closed circle: SD + PD (n = 9).
high pre-treatment serum VEGF level was a prognostic marker for treatment outcomes in metastatic colorectal, lung, and renal cell cancers, although it was not predictive for benefits of BEV-based treatment [20]. We found a decreasing trend of serum VEGF level by BEV treatment in this study, but VEGF was not a predictive factor of the curative effect of BEV treatment because there was no specific change between the two groups. Angiogenesis involves a series of component steps, including production and secretion of molecules, such as VEGF, by inflammatory and tumor cells in wound healing and inflammation. The response of VEGFinduced endothelial cells includes the release of proteases that degrade the basement membrane and infiltration into the surrounding ECM. Subsequent pro-
438
K. Hiura et al. / MMP-9/ANC score as a predictive biomarker for efficacy of bevacizumab plus platinum doublet chemotherapy
Progression-free survival (%)
100
high MMP-9/ANC group ( ≥ median) low MMP-9/ANC group ( < median)
80
60
40
20 0 0
100
200
300
400
500
600
700
800
Time (days)
A 100
high MMP-9/ANC group ( ≥ median) low MMP-9/ANC group ( < median)
Overall survival (%)
80
60
40
20 0
0
100
200
300
400
500
600
700
800
Time (days)
B
Fig. 5. Kaplan-Meier plots of progression-free survival and overall survival of patients treated with PCB chemotherapy. Patients were divided into a high scoring group (solid line, MMP-9/ANC median) and a low scoring group (dotted line, MMP-9/ANC < median) using the basal MMP-9/ANC scores. (A) Kaplan-Meier plots of progression-free survival: two groups were compared using a log-rank test (p = 0.032). (B) Kaplan-Meier plots of overall survival: two groups were compared using a log-rank test (p = 0.558).
cesses include the division and proliferation of endothelial cells in lumen formation, synthesis of a new basement membrane, and differentiation into blood vessels to increase blood flow. The main components of the basement membrane are type IV collagen, laminin, and proteoglycan, and it is thought that the basement membrane degradation process is a key mechanism in cancer invasion [8,21,22]. MMP2 and MMP-9 are type IV collagenases essential for degradation of the basement membrane. In this study, there were no significant differences in pre- and posttreatment serum MMP-2 and MMP-9 levels between or within groups. Although there were no significant differences in serum MMP-2 levels between lung cancer patients and healthy subjects, serum MMP-9 levels were significantly higher in the former which was consistent with the findings of Hrabec et al. [23]. Zymographic findings showed that serum MMP-9 displayed greater activity against gelatin than MMP-2 pre-treatment, and the activity of MMP-9 decreased throughout BEV treatment course. MMP-2 is mainly
produced by fibroblasts, while MMP-9 is mainly produced by neutrophils and is induced by inflammatory cytokines and growth factors [24–26]. High MMP-9 activity seems to be a consequence of inducible factors secreted by cancer cells. Therefore we corrected MMP-9 activity for the influence of ANC. The good responder group showed a significantly higher pre-treatment MMP-9/ANC score than the non-responder group, and there was a positive correlation between MMP-9/ANC score and the tumor reduction rates. Our results indicated that the treatment efficacy of PCB therapy will be high if the basal MMP9/ANC score is high. It seems that vascularization has considerable influence on tumor proliferation in size with high MMP-9/ANC scores and, subsequently, the curative effect of BEV will increase. The high MMP-9/ANC scoring group demonstrated a significant improvement in median PFS compared with the low MMP-9/ANC scoring group (636 vs. 196 days, respectively), but there was no improvement in median OS, indicating that the treatment efficacy of PCB therapy will be long lasting if the basal MMP-9/ANC score is high. Karihtala et al. reported that in BEV-treated ovarian cancer patients, MMP-9 expression was high more frequently among patients with PFS > 23 months than in those with PFS < 12 months [27], indicating that it is possible to predict not only the effect of the tumor reduction but also the effect of tumor stability by MMP-9 measurement. Currently, BEV is approved in Japan and other countries for the treatment of various cancers, including colorectal, lung, breast, kidney, and brain tumors, and approval for gastric and ovarian cancer, among others, is expected in the near future. In a preclinical model using human colon cancer xenografts, anti-tumor activity was observed that actually caused regression of the immature tumor vasculature by a single BEV infusion [28]. BEV normalizes existing tumor vasculature and reduces interstitial fluid pressure, which may help maximize the efficacy of concomitant cytotoxic chemotherapy [2,29,30]. It is important to identify cases sensitive to BEV treatment to enhance the curative effect of not only BEV but also the concomitant chemotherapeutic agents. In conclusion, the findings of the present study imply that the serum MMP-9/ANC score before PCB treatment may be a suitable predictive biomarker of the anti-tumor efficacy of PCB therapy. Careful setting of a cut-off score in a larger patient population will be needed to optimize the effectiveness of BEV for NSCLC treatment.
K. Hiura et al. / MMP-9/ANC score as a predictive biomarker for efficacy of bevacizumab plus platinum doublet chemotherapy
Conflict of interest The authors declare no conflict of interest.
Acknowledgements We thank Yuka Watanabe for her contribution to collect the samples and Kumiko Sugawara for her technical assistance.
References [1]
[2]
[3]
[4] [5]
[6]
[7]
[8]
[9]
[10]
[11]
L.G. Presta, H. Chen, S.J. O’Connor, V. Chisholm, Y.G. Meng, L. Krummen, M. Winkler and N. Ferrara, Humanization of an anti-vascular endothelial growth factor monoclonal antibody for the therapy of solid tumors and other disorders, Cancer Res 57 (1997), 4593-9. C.G. Willett, Y. Boucher, E. di Tomaso, D.G. Duda, L.L. Munn, R.T. Tong, D.C. Chung, D.V. Sahani, S.P. Kalva, S.V. Kozin, M. Mino, K.S. Cohen, D.T. Scadden, A.C. Hartford, A.J. Fischman, J.W. Clark, D.P. Ryan, A.X. Zhu, L.S. Blaszkowsky, H.X. Chen, P.C. Shellito, G.Y. Lauwers and R.K. Jain, Direct evidence that the VEGF-specific antibody bevacizumab has antivascular effects in human rectal cancer, Nat Med 10 (2004), 145-7. X. Zhu, S. Wu, W.L. Dahut and C.R. Parikh, Risks of proteinuria and hypertension with bevacizumab, an antibody against vascular endothelial growth factor: systematic review and meta-analysis, Am J Kidney Dis 49 (2007), 186-93. B.I. Lévy, Blood pressure as a potential biomarker of the efficacy angiogenesis inhibitor, Ann Oncol 20 (2009), 200-3. O. Mir, S. Ropert, J. Alexandre and F. Goldwasser, Hypertension as a surrogate marker for the activity of anti-VEGF agents, Ann Oncol 20 (2009), 967-70. S.E. Dahlberg, A.B. Sandler, J.R. Brahmer, J.H. Schiller and D.H. Johnson, Clinical course of advanced non-small-cell lung cancer patients experiencing hypertension during treatment with bevacizumab in combination with carboplatin and paclitaxel on ECOG 4599, J Clin Oncol 28 (2010), 949-54. A.M. Jubb and A.L. Harris, Biomarkers to predict the clinical efficacy of bevacizumab in cancer, Lancet Oncol 11 (2010), 1172-83. L.A. Liotta, K. Tryggvason, S. Garbisa, I. Hart, C.M. Foltz and S. Shafie, Metastatic potential correlates with enzymatic degradation of basement membrane collagen, Nature 284 (1980), 67-8. C. Johnson, H.J. Sung, S.M. Lessner, M.E. Fini and Z.S. Galis, Matrix metalloproteinase-9 is required for adequate angiogenic revascularization of ischemic tissues: potential role in capillary branching, Circ Res 94 (2004), 262-8. J. Fang, Y. Shing, D. Wiederschain, L. Yan, C. Butterfield, G. Jackson, J. Harper, G. Tamvakopoulos and M.A. Moses, Matrix metalloproteinase-2 is required for the switch to the angiogenic phenotype in a tumor model, Proc Natl Acad Sci USA 97 (2000), 3884-9. S. Silletti, T. Kessler, J. Goldberg, D.L. Boger and D.A. Cheresh, Disruption of matrix metalloproteinase 2 binding to integrin alpha vbeta 3 by an organic molecule inhibits angiogenesis and tumor growth in vivo, Proc Natl Acad Sci USA 98 (2001), 119-24.
[12]
439
P. Therasse, S.G. Arbuck, E.A. Eisenhauer, J. Wanders, R.S. Kaplan, L. Rubinstein, J. Verweij, M. Van Glabbeke, A.T. van Oosterom, M.C. Christian and S.G. Gwyther, 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 92 (2000), 205-16. [13] Y. Watanabe, R. Yamaguchi, S. Iwaki-Egawa, Y. Shimamori, Y. Fujimoto and H. Matsuno, Activation of progelatinase B in synovial fluids of patients with rheumatoid arthritis, with reference to stromelysin-1 and tissue inhibitor of matrix metalloproteinase-1, Clin Exp Rheumatol 17 (1999), 401-406. [14] T. Betsuyaku, M. Nishimura, K. Takeyabu, M. Tanino, P. Venge, S. Xu and Y. Kawakami, Neutrophil granule proteins in bronchoalveolar lavage fluid from subjects with subclinical emphysema, Am J Respir Crit Care Med 159 (1999), 1985-91. [15] E. van Cutsem, S. de Haas, Y.K. Kang, A. Ohtsu, N.C. Tebbutt, J. Ming Xu, W. Peng Yong, B. Langer, P. Delmar, S.J. Scherer and M.A. Shah, Bevacizumab in combination with chemotherapy as first-line therapy in advanced gastric cancer: a biomarker evaluation from the AVAGAST randomized phase III trial, J Clin Oncol 30 (2012), 2119-27. [16] S. Matsusaka, M. Suenaga, Y. Mishima, K. Takagi, Y. Terui, N. Mizunuma and K. Hatake, Circulating endothelial cells predict for response to bevacizumab-based chemotherapy in metastatic colorectal cancer, Cancer Chemother Pharmacol 68 (2011), 763-8. [17] H.I. Hurwitz, P.S. Douglas, J.P. Middleton, G.W. Sledge, D.H. Johnson, D.A. Reardon, D. Chen and O. Rosen, Analysis of early hypertension and clinical outcome with bevacizumab: results from seven phase III studies, Oncologist 18 (2013), 273-80. [18] J.P. Granger, Vascular endothelial growth factor inhibitors and hypertension: a central role for the kidney and endothelial factors?, Hypertension 54 (2009), 465-7. [19] A.M. Jubb, H.I. Hurwitz, W. Bai, E.B. Holmgren, P. Tobin, A.S. Guerrero, F. Kabbinavar, S.N. Holden, W.F. Novotny, G.D. Frantz, K.J. Hillan and H. Koeppen, Impact of vascular endothelial growth factor-A expression, thrombospondin2 expression, and microvessel density on the treatment effect of bevacizumab in metastatic colorectal cancer, J Clin Oncol 24 (2006), 217-27. [20] P.S. Hegde, A.M. Jubb, D. Chen, N.F. Li, G. Meng, C. Bernaards, R. Elliott, S. Scherer and D.S. Chen, Predictive Impact of Circulating Vascular Endothelial Growth Factor in 4 Phase III Trials Evaluating Bevacizumab, Clin Cancer Res 19 (2012), 929-37. [21] L.A. Liotta, P.S. Steeg and W.G. Stetler-Stevenson, Cancer metastasis and angiogenesis: an imbalance of positive and negative regulation, Cell 64 (1991), 327-36. [22] E.L. Rosenthal and L.M. Matrisian, Matrix metalloproteases in head and neck cancer, Head Neck 28 (2006), 639-48. [23] E. Hrabec, M. Strek, D. Nowak and Z. Hrabec, Elevated level of circulating matrix metalloproteinase-9 in patients with lung cancer, Respir Med 95 (2001), 1-4. [24] C. Trocme, P. Gaudin, S. Berthier, C. Barro, P. Zaoui and F. Morel, Human B lymphocytes synthesize the 92-kDa gelatinase, matrix metalloproteinase-9, J Biol Chem 273 (1998), 20677-84. [25] S. Okada, H. Kita, T.J. George, G.J. Gleich and K.M. Leiferman, Migration of eosinophils through basement membrane components in vitro: role of matrix metalloproteinase-9, Am J Respir Cell Mol Biol 17 (1997), 519-28.
440 [26] [27]
[28]
K. Hiura et al. / MMP-9/ANC score as a predictive biomarker for efficacy of bevacizumab plus platinum doublet chemotherapy A. Mauviel, Cytokine regulation of metalloproteinase gene expression, J Cell Biochem 53 (1993), 288-95. P. Karihtala, J. Maenpaa, T. Turpeenniemi-Hujanen and U. Puistola, Front-line bevacizumab in serous epithelial ovarian cancer: biomarker analysis of the FINAVAST trial, Anticancer Res 30 (2010), 1001-6. F. Yuan, Y. Chen, M. Dellian, N. Safabakhsh, N. Ferrara and R.K. Jain, Time-dependent vascular regression and permeability changes in established human tumor xenografts
[29]
[30]
induced by an anti-vascular endothelial growth factor/vascular permeability factor antibody, Proc Natl Acad Sci U S A 93 (1996), 14765-70. R.K. Jain, Normalizing tumor vasculature with antiangiogenic therapy: a new paradigm for combination therapy, Nat Med 7 (2001), 987-9. R.S. Kerbel, Antiangiogenic therapy: a universal chemosensitization strategy for cancer?, Science 312 (2006), 1171-5.
Cancer Biomarkers 15 (2015) 433–440 DOI 10.3233/CBM-150483 IOS Press
MMP-9/ANC score as a predictive biomarker for efficacy of bevacizumab plus platinum doublet chemotherapy in patients with advanced or recurrent non-squamous non-small cell lung cancer Kazuya Hiuraa,b , Akiko Shiraishia , Chinami Suzukic, Kei Takamurad , Makoto Yamamotod , Hitoshi Komorie , Yasuhiro Watanabea and Sachiko Iwaki-Egawaa,∗ a
Division of Life Science, Hokkaido Pharmaceutical University School of Pharmacy, Sapporo, Japan Department of Hospital Pharmacy, Abashiri Kosei General Hospital, Abashiri, Japan c Department of Hospital Pharmacy, Engaru Kosei General Hospital, Engaru, Japan d First Department of Internal Medicine, Obihiro Kosei General Hospital, Obihiro, Japan e Department of Hospital Pharmacy, Obihiro Kosei General Hospital, Obihiro, Japan b
Abstract. BACKGROUND: Bevacizumab is a recombinant humanized monoclonal antibody against vascular endothelial growth factor (VEGF), which is a key regulator of tumor angiogenesis. OBJECTIVE: To evaluate biomarkers to predict the benefit of paclitaxel and carboplatin plus bevacizumab (PCB) therapy in patients with advanced or recurrent non-squamous non-small cell lung cancer. METHODS: Among 21 patients treated with PCB, 10 were included in the good responder group and 11 in the non-responder group. Serum VEGF, MMP-2 and MMP-9 were measured using ELISA. RESULTS: There were no significant differences in these markers levels between groups. However, the good responder group showed a significantly higher pre-treatment MMP-9/ absolute neutrophil count (ANC) score than the non-responder group before the treatment (p = 0.014), and there was a positive correlation between the score and the tumor reduction rate (r = 0.57, p = 0.016). Furthermore, by dividing patients into a high scoring group (MMP-9/ANC median, n = 11) and a low scoring group (MMP-9/ANC < median, n = 10), former group showed a significant improvement in the median progression-free survival compared with latter group (636 vs. 196 days, p = 0.032). CONCLUSIONS: MMP-9/ANC score before PCB treatment may be a suitable biomarker to assess the anti-tumor effects of PCB therapy. Keywords: Absolute neutrophil count, matrix metalloproteinase-9, progression-free survival, tumor reduction rate, VEGF
1. Introduction ∗ Corresponding author: Sachiko Iwaki-Egawa, Division of Life Science, Hokkaido Pharmaceutical University, School of Pharmacy, 7-15-4-1 Maeda, Teine, Sapporo 006-8590, Hokkaido, Japan. Tel.: +81 11 676 8700; Fax: +81 11 676 8666; E-mail: sachiko@ hokuyakudai.ac.jp.
Bevacizumab (BEV) is a recombinant humanized monoclonal antibody against vascular endothelial growth factor (VEGF), which is a key regulator of tumor angiogenesis. BEV promotes its anti-angiogenic
c 2015 – IOS Press and the authors. All rights reserved ISSN 1574-0153/15/$35.00
434
K. Hiura et al. / MMP-9/ANC score as a predictive biomarker for efficacy of bevacizumab plus platinum doublet chemotherapy
effect by preventing soluble VEGF from binding to its specific receptors, thereby blocking and neutralizing VEGF-induced biological events [1,2]. As this mechanism of action is different from that of standard chemotherapeutic agents, BEV is not expected to cause typical chemotherapy-related toxicity or exacerbate the toxicity of concomitant chemotherapeutic agents. However, several principal BEV-related adverse events, such as arterial hypertension, proteinuria, and intestinal hemorrhage, have been observed in clinical trials; thus, oncologists must pay close attention to signs of these side effects [3]. Some recent reports have declared that BEV-induced hypertension may be a predictive marker for anti-angiogenic treatment efficacy [4–6]. A previous study reported that the occurrence of hypertension was associated with longer duration of the progression-free survival (PFS) and good overall survival (OS) compared with non-hypertensive patients with non-small cell lung cancer (NSCLC) receiving chemotherapy with paclitaxel and carboplatin plus BEV (PCB) [6]. There are, however, some problems with these findings. For example, it remains unclear whether a history of hypertension management can affect the predictive value of BEV treatment efficacy or whether the compensatory mechanisms to change blood pressure or multiple definitions of hypertension used in the study affected the results [7]. Therefore, we cannot definitively conclude that the onset of hypertension during PCB treatment is a clear predictor of improved outcome. VEGF binds to its receptors that are expressed on endothelial cell membranes, and produces and activates proteases, including matrix metalloproteinases (MMPs) [8], which increase vascular permeability by degrading the basement membrane and extracellular matrix (ECM), and promotes endothelial cells migration to the ECM and subsequent proliferation to induce new vascular grow. MMP-2 (72 kDa) and MMP-9 (92 kDa) are two MMP subtypes that are known to be involved in the degradation of type IV collagen, which composes the basement membrane of the capillary endothelium [9–11]. In the present study, we assessed the suitability of VEGF, MMP-2, MMP-9, and MMP-9/absolute neutrophil count (ANC) score as candidate biomarkers to predict the benefit of PCB chemotherapy in patients with advanced or recurrent non-squamous NSCLC.
2. Patients and methods 2.1. Patients Twenty-one patients in advanced or recurrent NSCLS were enrolled and treated with BEV plus platinum doublet combination chemotherapy on a 21-day treatment cycle from August 2010 to October 2011. According to the Response Evaluation Criteria in Solid Tumors (RECIST) guidelines [12], after the treatment patients were classified as patients with a complete response (CR), partial response (PR), stable disease (SD) or progressive disease (PD). A good responder group was deemed as the patients with CR or PR, and a nonresponder group was deemed as the patients with SD or PD. In this study, post-treatment means the time the evaluation was done. The primary endpoint of the study was response rate and tumor reduction rate. Secondary endpoints included PFS, OS, and toxicity. Concentrations of serum VEGF, MMP-2, and MMP-9 were measured at pre- and post-treatment with BEV and platinum doublet. As a control, healthy subjects were provided by the Hokkaido Red Cross Blood Center (Sapporo, Japan). Samples were stored at −30◦C or −80◦C until use. This study was performed in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of Obihiro Kosei General Hospital. Full, written, and informed consent was also obtained. 2.2. Enzyme-linked immunosorbent assay (ELISA) The following biomarkers were measured according to the manufacture’s instructions: VEGF (human VEGF Quantikine; R&D Systems, Minneapolis, MN, USA), MMP-2, and MMP-9 (human ELISA Biotrak; GE Healthcare, Buckinghamshire, UK). 2.3. Gelatin zymography Gelatinolytic activities of MMPs were analyzed using gelatin zymography [13]. Serum (equivalent to 0.5 µL) was run, without reduction, in a 0.1% SDS, 7.5% polyacrylamide gel impregnated with 0.3 mg/ml gelatin. Gel electrophoresis was performed at 4◦ C. After electrophoresis, the gel was washed in 2.5% Triton X-100 for 15 min and in 50 mM Tris-HCl, pH7.6 for 10 min. Then it was incubated overnight at 37◦ C in a buffer containing 30 mM Tris-HCl, pH 7.6, 5 mM CaCl2 , 0.2 M NaCl, and 0.02% NaN3 . Following this incubation, the gel was stained for proteins with 0.1%
K. Hiura et al. / MMP-9/ANC score as a predictive biomarker for efficacy of bevacizumab plus platinum doublet chemotherapy 3,000 2,500 VEGF (pg/mL)
Coomassie Blue R250 in 40% (v/v) isopropanol and destained in 7% (v/v) acetic acid. Gel images were acquired with a scanner, and MMP levels were quantified by densitometric analysis of the bands by Image J 1.47d (National Institutes of Health, USA).
435
2,000 1,500 1,000 500
2.4. Statistical analysis
0 CR + PR (n = 10)
CR + PR (n = 10)
SD + PD (n = 11)
P o st- t r e a t me n t
Fig. 1. Box-plots of serum VEGF levels in pre- and post-treatment with PCB chemotherapy. The ends of the whiskers represent the minimum and maximum of all of the data. Inbox bars show median for VEGF level of each group. Post-treatment means the time the evaluation was done. 10,000
3. Results
7,500
5,000
2,500
0 CR + PR (n = 10)
SD + PD (n = 11)
CR + PR (n = 10)
Pre -t reat me nt
SD + PD (n = 11)
Pos t-treatment
A 2,000
MMP-9 (ng/mL)
Of 21 patients who received BEV plus platinum doublet combination chemotherapy, 10 patients were classified as good responders (CR/PR, 1/9) and 11 as non-responders (SD/PD, 10/1). There were no significant differences between groups with regards to age, gender, performance status (PS), histology, clinical stage, mutations of the epidermal growth factor receptor gene (EGFR-mt), Brinkman index, or hypertension (i.e., high blood pressure by the end of cycle 1 was defined as blood pressure > 150/100-mmHg at any previous time or at least a 20-mmHg increase in diastolic blood pressure from baseline) (Table 1). There were no significant differences in serum VEGF levels between groups both pre- and posttreatment (Fig. 1). Pre-treatment VEGF levels were much higher in both groups than in healthy subjects (228 ± 144 pg/mL), and post-treatment, both decreased nearly to the levels of that observed in healthy subjects; however, there were no statistical differences between groups. Pre-treatment serum MMP-2 levels of both groups were almost the same as those of healthy subjects (2051 ± 228 ng/mL) and there were no significant differences in pre- and post-treatment serum MMP-2 levels between or within groups (Fig. 2A). Furthermore, there were no significant differences in serum MMP-9 levels, as with serum MMP-2 levels, between groups both pre- and post-treatment (Fig. 2B). How-
SD + PD (n = 11) P r e - tr e a t me n t
MMP-2 (ng/mL)
The data were analyzed with Student’s t-test, Welch’s test, Paired t-test in the parametric test, MannWhitney U test, Wilcoxon signed rank test in the nonparametric test, and Fisher’s exact probability test in the regression analysis. Spearman rank correlation coefficient was used for the analysis of the correlation. Survival probability was estimated using the Kaplan-Meier method, and the log-rank test was used to evaluate survival distributions among different patient groups. P values < 0.05 were considered statistically significant.
1,500
1,000
500
0 CR + PR (n = 10)
SD + PD (n = 11)
CR + PR (n = 10)
SD + PD (n = 11)
Post-treatment
Pre-treatment
B
Fig. 2. Box-plots of serum MMP-2 and MMP-9 levels in pre- and post-treatment with PCB chemotherapy. The ends of the whiskers represent the minimum and maximum of all of the data. Inbox bars show median for MMP-2 level (A) and MMP-9 level (B) of each group. Post-treatment means the time the evaluation was done.
ever, serum MMP-9 levels were significantly higher in both groups than in healthy subjects (11.7 ± 6.3 ng/mL), and the good responder group showed a tendency of higher pre-treatment MMP-9 levels than the non-responder group. Then, we compared serum MMP-2 and MMP-9 levels using gelatin zymography and ELISA. A typical
436
K. Hiura et al. / MMP-9/ANC score as a predictive biomarker for efficacy of bevacizumab plus platinum doublet chemotherapy Table 1 Patient characteristics Number of patients Age, range (median) Gender, males/females ECOG PS, 0/1/2 Histology, Ad/other Stage, IIIA/IIIB/IV EGFR-mt, +/−/unknown Brinkman index, range (median) Hypertension, +/−
CR + PR CR/PR (1/9) 46–77 (66) 5/5 4/4/2 8/2 0/2/8 2/8/0 0-1, 650 (80) 8/2
SD + PD SD/PD (10/1) 40–79 (57) 7/4 6/5/0 9/2 3/0/8 2/7/2 0-1, 200 (700) 7/4
P value 0.30a 0.42b 0.29c 0.67b 0.62c 0.67b 0.14a 0.37b
Abbreviations: CR = complete response; PR = partial response; SD = stable disease; PD = progressive disease; ECOG = Eastern Cooperative Oncology Group; PS = performance status; Ad = adenocarcinoma; EGFR-mt = epidermal growth factor receptor mutation. a Student’s t-test; b Fisher’s exact probability test; c Mann-Whitney’s U test.
zymographic pattern of serum collected from a patient who was classified as SD is shown in Fig. 3A. In zymography, only latent form bands of MMP-2 and MMP-9 (proMMP-2 and proMMP-9, respectively) were observed. The intensities of the MMP-9 bands decreased throughout the treatment course, though there was no significant change between pre- and posttreatment as we showed in Fig. 2B. The reason is that almost the evaluation was done at 2nd or 3rd treatment course and there was statistically no difference. In addition, the intensities of the MMP-2 bands remained virtually unchanged (Fig. 3A). Quantitation of the bands showed that MMP-9 before treatment had a 5.3-fold higher activity against gelatin than MMP-2. The serum MMP-2 and MMP-9 levels as determined by ELISA in this same patient throughout the treatment course are shown in Fig. 3B. The ELISA kits employed in this study recognize only the precursors of MMP2 and MMP-9. In addition, the ELISA results showed that MMP-9 levels were decreased and MMP-2 levels remained unchanged following treatment. The absolute amount of MMP-2 was, however, over 10-fold greater than that of MMP-9. The difference in intensities between zymography and ELISA may exist in the specific activity of both MMP-2 and MMP-9 against the gelatin. The results of other patient serum samples were almost the same (data not shown). Serum MMP-9 levels were corrected by ANC (Fig. 4A) because MMP-9 is produced by blood cells, such as neutrophils [14]. The good responder group showed a significantly higher pre-treatment MMP9/ANC score than the non-respoder group (p = 0.014), indicating that the treatment efficacy will be superior if the basal MMP-9/ANC score is high. There was a positive correlation (r = 0.57, p = 0.016) between the pre-treatment MMP-9/ANC scores and the tumor re-
duction rates of 19 patients (CR + PR, n = 10; SD + PD, n = 9) with primary tumor that could be measured after treatment (Fig. 4B). We divided the patients into a high scoring group (MMP-9/ANC median, n = 11) and a low scoring group (MMP-9/ANC < median, n = 10) and evaluated PFS and OS in each. The high MMP-9/ANC scoring group demonstrated a significant improvement in median PFS compared with the low MMP-9/ANC scoring group (636 vs. 196 days, p = 0.032) (Fig. 5A), but there was no improvement in median OS (p = 0.558) (Fig. 5B).
4. Discussion Several biomarkers have been reported as predictors of responsiveness to BEV therapy including plasma VEGF-A levels and tumor expression of neuropilin1 in gastric cancer [15], as well as the number of peripheral endothelial cells in colorectal cancer [16]. At present, however, these biomarkers have not yet been fully elucidated; therefore, we examined several biomarkers to predict the curative effects of BEV in advanced or recurrent NSCLC. No significant differences were found in age or gender, PS, histology, clinical stage, mutations of the EGFR-mt, or Brinkman index between the good responder and non-responder groups. There was also no significant difference between the two groups in the prevalence of early hypertension, which is reported to be a predictive marker of the curative effect of BEV treatment. Hurwitz et al. recently reported results in accordance with those of the present study [17]. However, some studies reported that the hypertension in response to VEGF inhibition was associated with a
K. Hiura et al. / MMP-9/ANC score as a predictive biomarker for efficacy of bevacizumab plus platinum doublet chemotherapy 1
2
3
4
5
437
P =0.014
6
proMMP-9
MMP-9/ANC score
200
150
100
50
proMMP-2 0 CR + PR (n = 10)
SD + PD (n = 11)
CR + PR (n = 10)
Pre-t re atm ent
SD + PD (n = 11) P ost-tr ea tm ent
A
A
100
300
5,000
4,000
200 3,000 150 2,000
MMP-2 ( , ng/mL)
MMP-9 ( , ng/mL)
250
Tumor reduction rate (%)
80
60
40
20
0
100
0
50
100
150
200
MMP-9/ANC score
1,000
50
0
0 Pretreatment 2nd
3rd
4th
5th
6th
Treatment course
B Fig. 3. MMP-2 and MMP-9 levels of serum collected from a patient who was classified as SD. (A) A typical zymographic pattern of serum. Lanes 1-6 contain an equivalent of 0.5 µL serum. Only latent form bands of MMP-2 and MMP-9 (proMMP-2, 72 kDa and proMMP-9, 92 kDa, respectively) were observed. Lane 1: pre-treatment, lane 2: before 2nd course treatment, lane 3: before 3rd course, lane 4: before 4th course, lane 5: before 5th course, and lane 6: before 6th course. (B) Concentrations of serum MMP-2 and MMP-9 of the same patient detected by ELISA. Open circle: MMP-9, closed circle: MMP-2.
significant reduction in nitric oxide (NO) production, which is caused by inhibition of NO synthase as a result of endothelial cell injury [18]. In our preliminary study, we also measured serum NO levels, but found no significant differences between groups pre- and posttreatment (data not shown). Jubb et al. reported that the level of VEGF expression was neither a prognostic nor a predictive marker of benefit from BEV treatment in colon cancer [19]. Hegde et al. also indicated that a
B
Fig. 4. MMP-9/ANC score in pre- and post-treatment and correlation between the tumor reduction rates and the pretreatment MMP-9/ANC scores. (A) Box-plots of MMP-9/ANC score in pre- and post-treatment with PCB chemotherapy. The ends of the whiskers represent the minimum and maximum of all of the data. Inbox bars show median for MMP-9/ANC score of each group. (B) Correlation between the tumor reduction rates and the pre-treatment MMP-9/ANC scores. A line represents the linear regression (r = 0.57, p < 0.05). Open circle: CR + PR (n = 10), closed circle: SD + PD (n = 9).
high pre-treatment serum VEGF level was a prognostic marker for treatment outcomes in metastatic colorectal, lung, and renal cell cancers, although it was not predictive for benefits of BEV-based treatment [20]. We found a decreasing trend of serum VEGF level by BEV treatment in this study, but VEGF was not a predictive factor of the curative effect of BEV treatment because there was no specific change between the two groups. Angiogenesis involves a series of component steps, including production and secretion of molecules, such as VEGF, by inflammatory and tumor cells in wound healing and inflammation. The response of VEGFinduced endothelial cells includes the release of proteases that degrade the basement membrane and infiltration into the surrounding ECM. Subsequent pro-
438
K. Hiura et al. / MMP-9/ANC score as a predictive biomarker for efficacy of bevacizumab plus platinum doublet chemotherapy
Progression-free survival (%)
100
high MMP-9/ANC group ( ≥ median) low MMP-9/ANC group ( < median)
80
60
40
20 0 0
100
200
300
400
500
600
700
800
Time (days)
A 100
high MMP-9/ANC group ( ≥ median) low MMP-9/ANC group ( < median)
Overall survival (%)
80
60
40
20 0
0
100
200
300
400
500
600
700
800
Time (days)
B
Fig. 5. Kaplan-Meier plots of progression-free survival and overall survival of patients treated with PCB chemotherapy. Patients were divided into a high scoring group (solid line, MMP-9/ANC median) and a low scoring group (dotted line, MMP-9/ANC < median) using the basal MMP-9/ANC scores. (A) Kaplan-Meier plots of progression-free survival: two groups were compared using a log-rank test (p = 0.032). (B) Kaplan-Meier plots of overall survival: two groups were compared using a log-rank test (p = 0.558).
cesses include the division and proliferation of endothelial cells in lumen formation, synthesis of a new basement membrane, and differentiation into blood vessels to increase blood flow. The main components of the basement membrane are type IV collagen, laminin, and proteoglycan, and it is thought that the basement membrane degradation process is a key mechanism in cancer invasion [8,21,22]. MMP2 and MMP-9 are type IV collagenases essential for degradation of the basement membrane. In this study, there were no significant differences in pre- and posttreatment serum MMP-2 and MMP-9 levels between or within groups. Although there were no significant differences in serum MMP-2 levels between lung cancer patients and healthy subjects, serum MMP-9 levels were significantly higher in the former which was consistent with the findings of Hrabec et al. [23]. Zymographic findings showed that serum MMP-9 displayed greater activity against gelatin than MMP-2 pre-treatment, and the activity of MMP-9 decreased throughout BEV treatment course. MMP-2 is mainly
produced by fibroblasts, while MMP-9 is mainly produced by neutrophils and is induced by inflammatory cytokines and growth factors [24–26]. High MMP-9 activity seems to be a consequence of inducible factors secreted by cancer cells. Therefore we corrected MMP-9 activity for the influence of ANC. The good responder group showed a significantly higher pre-treatment MMP-9/ANC score than the non-responder group, and there was a positive correlation between MMP-9/ANC score and the tumor reduction rates. Our results indicated that the treatment efficacy of PCB therapy will be high if the basal MMP9/ANC score is high. It seems that vascularization has considerable influence on tumor proliferation in size with high MMP-9/ANC scores and, subsequently, the curative effect of BEV will increase. The high MMP-9/ANC scoring group demonstrated a significant improvement in median PFS compared with the low MMP-9/ANC scoring group (636 vs. 196 days, respectively), but there was no improvement in median OS, indicating that the treatment efficacy of PCB therapy will be long lasting if the basal MMP-9/ANC score is high. Karihtala et al. reported that in BEV-treated ovarian cancer patients, MMP-9 expression was high more frequently among patients with PFS > 23 months than in those with PFS < 12 months [27], indicating that it is possible to predict not only the effect of the tumor reduction but also the effect of tumor stability by MMP-9 measurement. Currently, BEV is approved in Japan and other countries for the treatment of various cancers, including colorectal, lung, breast, kidney, and brain tumors, and approval for gastric and ovarian cancer, among others, is expected in the near future. In a preclinical model using human colon cancer xenografts, anti-tumor activity was observed that actually caused regression of the immature tumor vasculature by a single BEV infusion [28]. BEV normalizes existing tumor vasculature and reduces interstitial fluid pressure, which may help maximize the efficacy of concomitant cytotoxic chemotherapy [2,29,30]. It is important to identify cases sensitive to BEV treatment to enhance the curative effect of not only BEV but also the concomitant chemotherapeutic agents. In conclusion, the findings of the present study imply that the serum MMP-9/ANC score before PCB treatment may be a suitable predictive biomarker of the anti-tumor efficacy of PCB therapy. Careful setting of a cut-off score in a larger patient population will be needed to optimize the effectiveness of BEV for NSCLC treatment.
K. Hiura et al. / MMP-9/ANC score as a predictive biomarker for efficacy of bevacizumab plus platinum doublet chemotherapy
Conflict of interest The authors declare no conflict of interest.
Acknowledgements We thank Yuka Watanabe for her contribution to collect the samples and Kumiko Sugawara for her technical assistance.
References [1]
[2]
[3]
[4] [5]
[6]
[7]
[8]
[9]
[10]
[11]
L.G. Presta, H. Chen, S.J. O’Connor, V. Chisholm, Y.G. Meng, L. Krummen, M. Winkler and N. Ferrara, Humanization of an anti-vascular endothelial growth factor monoclonal antibody for the therapy of solid tumors and other disorders, Cancer Res 57 (1997), 4593-9. C.G. Willett, Y. Boucher, E. di Tomaso, D.G. Duda, L.L. Munn, R.T. Tong, D.C. Chung, D.V. Sahani, S.P. Kalva, S.V. Kozin, M. Mino, K.S. Cohen, D.T. Scadden, A.C. Hartford, A.J. Fischman, J.W. Clark, D.P. Ryan, A.X. Zhu, L.S. Blaszkowsky, H.X. Chen, P.C. Shellito, G.Y. Lauwers and R.K. Jain, Direct evidence that the VEGF-specific antibody bevacizumab has antivascular effects in human rectal cancer, Nat Med 10 (2004), 145-7. X. Zhu, S. Wu, W.L. Dahut and C.R. Parikh, Risks of proteinuria and hypertension with bevacizumab, an antibody against vascular endothelial growth factor: systematic review and meta-analysis, Am J Kidney Dis 49 (2007), 186-93. B.I. Lévy, Blood pressure as a potential biomarker of the efficacy angiogenesis inhibitor, Ann Oncol 20 (2009), 200-3. O. Mir, S. Ropert, J. Alexandre and F. Goldwasser, Hypertension as a surrogate marker for the activity of anti-VEGF agents, Ann Oncol 20 (2009), 967-70. S.E. Dahlberg, A.B. Sandler, J.R. Brahmer, J.H. Schiller and D.H. Johnson, Clinical course of advanced non-small-cell lung cancer patients experiencing hypertension during treatment with bevacizumab in combination with carboplatin and paclitaxel on ECOG 4599, J Clin Oncol 28 (2010), 949-54. A.M. Jubb and A.L. Harris, Biomarkers to predict the clinical efficacy of bevacizumab in cancer, Lancet Oncol 11 (2010), 1172-83. L.A. Liotta, K. Tryggvason, S. Garbisa, I. Hart, C.M. Foltz and S. Shafie, Metastatic potential correlates with enzymatic degradation of basement membrane collagen, Nature 284 (1980), 67-8. C. Johnson, H.J. Sung, S.M. Lessner, M.E. Fini and Z.S. Galis, Matrix metalloproteinase-9 is required for adequate angiogenic revascularization of ischemic tissues: potential role in capillary branching, Circ Res 94 (2004), 262-8. J. Fang, Y. Shing, D. Wiederschain, L. Yan, C. Butterfield, G. Jackson, J. Harper, G. Tamvakopoulos and M.A. Moses, Matrix metalloproteinase-2 is required for the switch to the angiogenic phenotype in a tumor model, Proc Natl Acad Sci USA 97 (2000), 3884-9. S. Silletti, T. Kessler, J. Goldberg, D.L. Boger and D.A. Cheresh, Disruption of matrix metalloproteinase 2 binding to integrin alpha vbeta 3 by an organic molecule inhibits angiogenesis and tumor growth in vivo, Proc Natl Acad Sci USA 98 (2001), 119-24.
[12]
439
P. Therasse, S.G. Arbuck, E.A. Eisenhauer, J. Wanders, R.S. Kaplan, L. Rubinstein, J. Verweij, M. Van Glabbeke, A.T. van Oosterom, M.C. Christian and S.G. Gwyther, 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 92 (2000), 205-16. [13] Y. Watanabe, R. Yamaguchi, S. Iwaki-Egawa, Y. Shimamori, Y. Fujimoto and H. Matsuno, Activation of progelatinase B in synovial fluids of patients with rheumatoid arthritis, with reference to stromelysin-1 and tissue inhibitor of matrix metalloproteinase-1, Clin Exp Rheumatol 17 (1999), 401-406. [14] T. Betsuyaku, M. Nishimura, K. Takeyabu, M. Tanino, P. Venge, S. Xu and Y. Kawakami, Neutrophil granule proteins in bronchoalveolar lavage fluid from subjects with subclinical emphysema, Am J Respir Crit Care Med 159 (1999), 1985-91. [15] E. van Cutsem, S. de Haas, Y.K. Kang, A. Ohtsu, N.C. Tebbutt, J. Ming Xu, W. Peng Yong, B. Langer, P. Delmar, S.J. Scherer and M.A. Shah, Bevacizumab in combination with chemotherapy as first-line therapy in advanced gastric cancer: a biomarker evaluation from the AVAGAST randomized phase III trial, J Clin Oncol 30 (2012), 2119-27. [16] S. Matsusaka, M. Suenaga, Y. Mishima, K. Takagi, Y. Terui, N. Mizunuma and K. Hatake, Circulating endothelial cells predict for response to bevacizumab-based chemotherapy in metastatic colorectal cancer, Cancer Chemother Pharmacol 68 (2011), 763-8. [17] H.I. Hurwitz, P.S. Douglas, J.P. Middleton, G.W. Sledge, D.H. Johnson, D.A. Reardon, D. Chen and O. Rosen, Analysis of early hypertension and clinical outcome with bevacizumab: results from seven phase III studies, Oncologist 18 (2013), 273-80. [18] J.P. Granger, Vascular endothelial growth factor inhibitors and hypertension: a central role for the kidney and endothelial factors?, Hypertension 54 (2009), 465-7. [19] A.M. Jubb, H.I. Hurwitz, W. Bai, E.B. Holmgren, P. Tobin, A.S. Guerrero, F. Kabbinavar, S.N. Holden, W.F. Novotny, G.D. Frantz, K.J. Hillan and H. Koeppen, Impact of vascular endothelial growth factor-A expression, thrombospondin2 expression, and microvessel density on the treatment effect of bevacizumab in metastatic colorectal cancer, J Clin Oncol 24 (2006), 217-27. [20] P.S. Hegde, A.M. Jubb, D. Chen, N.F. Li, G. Meng, C. Bernaards, R. Elliott, S. Scherer and D.S. Chen, Predictive Impact of Circulating Vascular Endothelial Growth Factor in 4 Phase III Trials Evaluating Bevacizumab, Clin Cancer Res 19 (2012), 929-37. [21] L.A. Liotta, P.S. Steeg and W.G. Stetler-Stevenson, Cancer metastasis and angiogenesis: an imbalance of positive and negative regulation, Cell 64 (1991), 327-36. [22] E.L. Rosenthal and L.M. Matrisian, Matrix metalloproteases in head and neck cancer, Head Neck 28 (2006), 639-48. [23] E. Hrabec, M. Strek, D. Nowak and Z. Hrabec, Elevated level of circulating matrix metalloproteinase-9 in patients with lung cancer, Respir Med 95 (2001), 1-4. [24] C. Trocme, P. Gaudin, S. Berthier, C. Barro, P. Zaoui and F. Morel, Human B lymphocytes synthesize the 92-kDa gelatinase, matrix metalloproteinase-9, J Biol Chem 273 (1998), 20677-84. [25] S. Okada, H. Kita, T.J. George, G.J. Gleich and K.M. Leiferman, Migration of eosinophils through basement membrane components in vitro: role of matrix metalloproteinase-9, Am J Respir Cell Mol Biol 17 (1997), 519-28.
440 [26] [27]
[28]
K. Hiura et al. / MMP-9/ANC score as a predictive biomarker for efficacy of bevacizumab plus platinum doublet chemotherapy A. Mauviel, Cytokine regulation of metalloproteinase gene expression, J Cell Biochem 53 (1993), 288-95. P. Karihtala, J. Maenpaa, T. Turpeenniemi-Hujanen and U. Puistola, Front-line bevacizumab in serous epithelial ovarian cancer: biomarker analysis of the FINAVAST trial, Anticancer Res 30 (2010), 1001-6. F. Yuan, Y. Chen, M. Dellian, N. Safabakhsh, N. Ferrara and R.K. Jain, Time-dependent vascular regression and permeability changes in established human tumor xenografts
[29]
[30]
induced by an anti-vascular endothelial growth factor/vascular permeability factor antibody, Proc Natl Acad Sci U S A 93 (1996), 14765-70. R.K. Jain, Normalizing tumor vasculature with antiangiogenic therapy: a new paradigm for combination therapy, Nat Med 7 (2001), 987-9. R.S. Kerbel, Antiangiogenic therapy: a universal chemosensitization strategy for cancer?, Science 312 (2006), 1171-5.
