Ann Nucl Med DOI 10.1007/s12149-015-0975-5
ORIGINAL ARTICLE
99m
Tc-3PRGD2 SPECT/CT predicts the outcome of advanced nonsquamous non-small cell lung cancer receiving chemoradiotherapy plus bevacizumab
Qingjie Ma1 • Kaiyin Min1 • Ting Wang1 • Bin Chen1 • Qiang Wen1 Fan Wang2 • Tiefeng Ji1 • Shi Gao1
•
Received: 11 March 2015 / Accepted: 12 April 2015 Ó The Japanese Society of Nuclear Medicine 2015
Abstract Background Functional imaging can help clinicians assess the individual response of advanced nonsquamous non-small cell lung cancer (NSCLC) to chemoradiation therapy plus bevacizumab. Our purpose is to investigate the ability of 99mTc-3PRGD2 single photon emission computed tomography/computed tomography (SPECT/CT) in predicting the early response to treatment. Methods Patients with advanced nonsquamous NSCLC diagnosed by histological or cytological examination were imaged with 99mTc-3PRGD2 SPECT/CT at 3 time points: 1–3 days before the start of treatment (SPECT1), 40 Gy radiotherapy with 2 cycles of chemotherapy plus bevacizumab (SPECT2) and 4 weeks after chemoradiotherapy plus bevacizumab (SPECT3). The images were evaluated semiquantitatively by measuring the tumor to non-tumor ratio (T/N) and calculating the percentage change in T/N ratio. Short-term outcome was assessed by the treatment response evaluation according to the Response Evaluation Criteria in Solid Tumors criteria as: complete response (CR), partial response (PR), stable disease (SD) and progressive disease (PD). Patients were divided two groups: responders (CR and PR) and nonresponders (SD and PD). To determine a threshold for percent reduction in T/N ratios, receiver-operating characteristic (ROC) curve analysis & Tiefeng Ji [email protected] & Shi Gao [email protected] 1
Department of Nuclear Medicine, China-Japan Union Hospital of Jilin University, Changchun, China
2
Medical Isotopes Research Center, Peking University, Beijing, China
was used. Patients were grouped again based on the threshold of P1 (the change percentage from SPECT1 to SPECT2) and P2 (the change percentage from SPECT1 to SPECT3): P1 responders and P1 nonresponders; P2 responders and P2 nonresponders. Patients were followed up starting 4 weeks after completion of therapy and then every 3 months for the first 2 years and every 6 months after 2 years. OS of P1 responders, P1 nonresponders, P2 responders and P2 nonresponders was estimated and graphically illustrated using the Kaplan–Meier method and the log-rank test was used to test the null hypotheses of equal OS in subgroups of patients. Results A total of 28 patients completed all imaging and treatment. All primary lung tumors were well visualized on SPECT1. The mean T/N ratio of SPECT1 in responders and nonresponders was not statistically different (2.73 ± 0.59 vs. 2.59 ± 0.52, p [ 0.05). At SPECT2 and SPECT3, the mean T/N ratios were both lower in the responders compared with the nonresponders and had statistical significance (p \ 0.05). P1 and P2 in the responders was larger than the nonresponders with significant difference (P1: 34.18 ± 21.55 % vs. 9.02 ± 14.02 %, p \ 0.05; P2: 53.02 ± 15.50 % vs. 7.74 ± 37.95 %, p \ 0.05). The optimal threshold of P1 that can discriminate between P1 responders and P1 nonresponders was greater than 25.9 % reduction, and that of P2 that can discriminate between P2 responders and P2 nonresponders was 34.0 % reduction. The area under the ROC curve (AUC) of P1 and P2 for determining residual disease was 0.856 and 0.909, respectively; but there was no statistical significance between them (p [ 0.05). There was a significant difference for OS between P1 responders and P1 nonresponders (p \ 0.05), and also for OS between P2 responders and P2 nonresponders (p \ 0.05). But there was no difference between the P1 responders and P2 responders (p [ 0.05), or
123
Ann Nucl Med
between the P1 nonresponders and P2 nonresponders (p [ 0.05). Conclusion A 99mTc-3PRGD2 SPECT/CT after two cycles of chemoradiotherapy plus bevacizumab can predict patients who will have a better response to treatment and survival. Keywords 99mTc-3PRGD2 SPECT/CT Lung cancer Response Prediction Bevacizumab
by histological or cytological examination were enrolled in this study at China–Japan Union Hospital, Changchun, China. These patients are not indicated for surgery due to metastases or other co-morbidities. The tumors were measurable with positive imaging according to Response Evaluation Criteria in Solid Tumors (RECIST). Informed written consent to participate in this study was obtained from all patients. The study protocol was approved by the ethics committee and the Institutional Review Board of China–Japan Union Hospital, Changchun, China.
Introduction
Treatment
Lung cancer ranks as one of the most common types of cancer in China and worldwide with an increasing incidence rate. It is the leading cause of malignant disease mortality [1, 2]. Nearly 80 % of lung cancer cases are nonsmall cell lung cancer (NSCLC), and the overall five-year survival of such cases can barely reach 15 % [3–5]. Although great advances have been achieved in diagnostics and treatment, metastatic disease still occurs in 70 % of NSCLC patients after receiving treatment [6, 7]. Therefore, accurate methods that enable outcome prediction to screen patients who are likely to benefit from treatment are required urgently. Functional imaging, such as positron emission tomography/computed tomography (PET/CT) and single photon emission computed tomography/CT (SPECT/CT), can both help clinicians design individualized treatment to improve the outcome of advanced NSCLC [8–10]. Novel avb3specific tracers for SPECT, like technetium-99m (99mTc) radiolabeled Arg-Gly-Asp (RGD) peptides and analogs, can specifically target the integrin avb3, which plays a critical role in the regulation of tumor angiogenesis and metastasis [11, 12]. Therefore, such tracers have been intensively investigated for noninvasive functional imaging of tumors in pre-clinical and clinical studies. 99mTc3PRGD2 SPECT/CT imaging can provide more accurate and timelier information than structural imaging alone [13– 15]. Therefore, we conducted this prospective study to investigate the correlation between different tumor angiogenesis changes with the end-of-treatment response to determine whether SPECT can predict short-term outcomes of chemoradiotherapy plus bevacizumab in patients with advanced nonsquamous NSCLC.
Patients received concomitant chemoradiotherapy plus bevacizumab designed by medical oncologists according to international standards. The chemotherapy regimen used in this study was cisplatin (administered i.v. at 80 mg/m2 on day 1) and gemcitabine (i.v. at 1250 mg/m2 on days 1 and 8), repeated every 3 weeks for up to six cycles. Bevacizumab was administered i.v. at 15 mg/kg concurrently with chemotherapy every 3 weeks on day 1. The radiotherapy regimen was intensity-modulated radiotherapy (IMRT). The patients were treated with a conventional fractionation scheme 5 days per week with a fractional dose of 5 9 2 Gy per week and late course accelerated hyperfractionated RT with 1.4 Gy twice daily up to a total cumulative dose of 63–67.5 Gy. RT was based on a planning CT image. The gross tumor volume (GTV) was delineated to the primary tumor and metastatic regional lymph nodes, and the planning target volume included the GTV with a margin of 1.0–1.5 cm.
Materials and methods Patients From May 2011 to January 2014,patients with nonsquamous advanced NSCLC (IIIB or IV tumor stage) diagnosed
123
99m
99m
Tc-3PRGD2 SPECT/CT imaging
Tc-3PRGD2 SPECT/CT imaging was performed at three time points: SPECT1 (1–3 days before the start of treatment), SPECT2 (40 Gy IMRT with 2 cycles of chemotherapy plus bevacizumab) and SPECT3 (4 weeks after completion of treatment) using a SPECT/CT scanner (Philips Precedence). Radiolabeling and quality control procedures for 3PRGD2 were performed as described previously [14]. 3PRGD2 was radiolabeled with 378 ± 86 MBq 99m technetium and thereafter administered via a single intravenous bolus injection in the contralateral arm to the affected lung, followed by a 10 mL saline flush. During imaging, patients were in the supine position with raised arms. X-ray scanning by a double-head c camera was performed to show the detecting range. The following CT scanning was set at a matrix of 256 9 256 and a 5-mm layer thickness. After collecting CT images, the detecting bench was positioned automatically for
Ann Nucl Med
SPECT data collection. The matrix was 128 9 128 pixels, and the photopeak was centered at 140 keV with a symmetrical 20 % window. Imaging with both radiotracers was performed using 6° angular steps in a 20 s time frame. The distance between the chest and detector was minimized. The Digital Imaging and Communications in Medicine image files of each patient were saved on optical disks and transferred to Extended BrillianceTM workspace (Philips Healthcare) for centralized reconstruction, reading and analysis. Imaging analysis All images were visually analyzed by two experienced nuclear medicine physicians without knowledge of the patients’ history. For semiquantitative analysis of tracer uptake in tumors following SPECT1, SPECT2 and SPECT3, regions of interest (ROIs) were drawn around the entire tumor and to the contralateral normal lung tissue with the same size of the tumor, and tumor to non-tumor localization (T/N) ratios of 99mTc-3PRGD2 SPECT/CT were determined. We used T/N ratios for further quantitative assessment and potential differentiation of lesions. The percentage changes in T/N ratios between SPECT1 and SPECT2 (P1) or SPECT3 (P2) were calculated using the following formulae: P1 ¼ ½T=NSPECT1 T=NSPECT2 =T=NSPECT1 100 %; P2 ¼ ½T=NSPECT1 T=NSPECT3 =T=NSPECT1 100 %: Clinical response validation Clinical short-term outcome was verified using CT of SPECT3 according to RECIST criteria as: complete response (CR), partial response (PR), stable disease (SD) and progressive disease (PD). Patients were divided into two groups: responders (CR and PR) and nonresponders (SD and PD). Patient follow-up Patients were followed up starting at 4 weeks after completion of therapy and then every 3 months for the first 2 years and every 6 months after 2 years. The follow-up includes history taking and physical examination. Overall survival (OS) was defined as the time from the completion of treatment to death from any cause. Statistical analysis All data were expressed as the mean ± standard deviation (SD). Statistical analyses were performed using software
SPSS for Windows (version 19.0). The differences in the dosimetric parameters between responders and nonresponders were evaluated by t test. To determine a threshold for percent reduction in T/N ratios, receiver-operating characteristic (ROC) curve analysis was used. The area under the ROC curve (AUC) provided the predictive power for 99mTc-3PRGD2 SPECT/CT imaging. Patients were grouped again based on the threshold of P1 and P2: P1 responders and P1 nonresponders; P2 responders and P2 nonresponders. OS of P1 responders, P1 nonresponders, P2 responders and P2 nonresponders was estimated and graphically illustrated using the Kaplan–Meier method, and the log-rank test was used to test the null hypotheses of equal OS in subgroups of patients. p \ 0.05 was considered to indicate statistical significance.
Results Patients Our original plan was to recruit thirty patients for this study, but one patient voluntarily dropped out and one patient lost contact. These two cases were excluded from further analysis. The remaining 28 patients completed all 3 SPECT/CT phases and treatment. Patient characteristics are shown in Table 1. The mean age was 55.1 ± 13.2 years (range 32–67 years). The mean followup time for all patients was 15 months (3–30 months).
Table 1 Characteristics of 28 patients enrolled in the study Characteristic
Number
Sex Male
12
Female
16
Mean age ± SD
55.1 ± 13.2
Tumor size, pretreatment (CT), mm ± SD
58.5 ± 14.7
Clinical stage IIIB
22
IV
6
Pathological type Adenocarcinoma
24
Large cell carcinoma Mucoepidermoid carcinoma
3 1
Clinical response Complete response
5
Partial response
12
Stable disease
8
Progressive disease
3
123
Ann Nucl Med
Clinical response Based on RECIST, 5 patients achieved CR and 12 had PR. The overall response rate was 60.71 %. By comparison, 8 patients attained SD and only 3 patients had PD. T/N ratios All primary lung tumors were well visualized on SPECT1 and evaluated using T/N ratios. The absolute T/N ratios and change in T/N ratios of three time points determined by two observers are summarized in Table 2. The mean T/N ratio of SPECT1 in all patients was 2.68 ± 0.56. No significant difference was observed in the mean pre-therapy T/N ratio between responders and nonresponders (2.73 ± 0.59 vs. 2.59 ± 0.52, p [ 0.05). At SPECT2 and SPECT3, the mean T/N ratios were both lower in the responders compared with the nonresponders and had statistical significance (SPECT2: 1.72 ± 0.46 vs. 2.33 ± 0.46, p \ 0.05; SPECT3: 1.23 ± 0.31 vs. 2.30 ± 0.74, p \ 0.05). P1 in the responders was larger than the nonresponders with significant difference (34.18 ± 21.55 vs. 9.02 ± 14.02, p \ 0.05). The P2 between the two groups also had a significant difference (53.02 ± 15.50 vs. 7.74 ± 37.95, p \ 0.05). Figure 1 shows typical examples of 99mTc-3PRGD2 SPECT/CT in patients with responding and nonresponding tumors. Figure 2 shows the different patterns of changes in T/N ratios of the responders and nonresponders at three time points. The responders showed a larger initial drop in T/N ratio at SPECT2 from SPECT1, whereas the nonresponders showed a minimal drop in the ratio and even higher values. The three outliers in Fig. 2 demonstrated an unexpected change in the pattern of uptake between SPECT2 and SPECT3. The first outlier was in the CR group. During SPECT2, the outlier’s T/N ratio demonstrated a poor response to treatment (10.6 % reduction in uptake), but subsequently showed a 51.4 % reduction in T/N ratio at SPECT3 with no visual image on the final SPECT image. The second outlier was a PR patient. There was an early 5.6 % increase in the T/N ratio from SPECT1 to SPECT2, but a 12.2 % reduction in T/N ratio from SPECT1 to SPECT3. In the final CT, the patient was found to have
Table 2 T/N ratios and its changes in three time points of SPECT/CT
T/N ratio
All patients
mild residual disease. The third outlier was a PD patient with a marked reduction from SPECT1 to SPECT3 (57.3 % reduction in uptake). But the lesion became larger, as identified by the final CT. Figure 3 illustrates P1 and P2 according to RECIST groupings. As seen in Fig. 3, the thresholds at P1 and P2 were 25.9 and 34.0 % drop change. ROC curve analysis The optimal threshold of P1 that can discriminate between P1 responders and P1 nonresponders was greater than 25.9 % reduction, and the optimal threshold of P2 that can discriminate between P2 responders and P2 nonresponders was 34.0 % reduction. The AUC of the ROCs for P1 and P2 was 0.856 and 0.909, respectively (Fig. 4). The predicting ability of P2 was higher than P1, but without statistical significance (p [ 0.05). Tumor response was a significant predictor of OS in the analysis. The one-year cumulative survival rates for P1 responders, P1 nonresponders, P2 responders and P2 nonresponders were 91.7, 56.3, 82.4 and 63.6 %. There was a significant difference for OS between P1 responders and P1 nonresponders (p \ 0.05), and also for OS between P2 responders and P2 nonresponders (p \ 0.05; Fig. 5). But there was no difference between the P1 responders and P2 responders (p [ 0.05), or between the P1 nonresponders and P2 nonresponders (p [ 0.05).
Discussion Patients with advanced NSCLC have a higher risk of cancer-related death in comparison with less advanced stages of the cancer. Concomitant chemoradiotherapy is widely used for first-line treatment of advanced NSCLC. Bevacizumab is an anti-vascular endothelial growth factor mAb with good clinical benefits established in a variety of tumors, including lung cancer [16]. So in nonsquamous NSCLC, bevacizumab is always used with concomitant chemoradiotherapy on the basis of its favorable efficacy and tolerability profile. However, the response to this therapeutic regimen is variable [17, 18]. Previous studies
Responder group*
Nonresponder group*
SPECT1
2.68 ± 0.56
2.73 ± 0.59
2.59 ± 0.52
0.527
SPECT2
1.96 ± 0.55
1.72 ± 0.46
2.33 ± 0.46
0.002
SPECT3
1.65 ± 0.74
1.23 ± 0.31
2.30 ± 0.74
0.001
P1 (%)
24.30 ± 22.46
34.18 ± 21.55
9.02 ± 14.02
0.001
P2 (%)
35.23 ± 34.39
53.02 ± 15.50
7.74 ± 37.95
0.003
p* is the comparative value of Responder group and Nonresponder group
123
p*
Ann Nucl Med
Fig. 1 Two typical examples of 99mTc-3PRGD2 SPECT/CT images in responders (a) and nonresponders (b) acquired at SPECT1, SPECT2 and SPECT3. For the responder, the mass size decreased significantly and the mass was almost invisible during the final
SPECT/CT (red arrows). For the nonresponder, necrosis appeared at the center of the lesion (blue arrows), and pathologic fractures from bone metastases (pink arrow) appeared at the same time (color figure online)
Fig. 2 The changes in the T/N ratio of the two groups (4 subgroups) at the three time points
123
Ann Nucl Med
Fig. 3 Percent changes in the T/N ratio of
99m
Tc-3PRGD2 of P1 (a) and P2 (b) of the responders versus nonresponders
Fig. 4 Comparison between ROC analysis of P1 and P2
have shown that patients achieving a good response to therapeutic regimens have a better prognosis than those with poor response [18, 19]. Long-term ineffective treatment not only brings great pain to patients, but also increases their economic burden. Therefore, there is an urgent need to evaluate the effect of therapy in a timely manner. Unfortunately, the limitations of conventional tools (X-ray, CT) to measure response make it difficult to accurately assess the quantity of residual viable tumor over the course of treatment [18]. Molecular imaging can provide information about the changes of the tumor on the molecular level and quantify the amount of viable residual disease over the course of treatment. Based on the response to therapy, clinicians can make individualized treatment
123
Fig. 5 OS of P1 responders, P2 responders, P1 nonresponders and P2 nonresponders was estimated by the Kaplan–Meier method
adjustments quickly. For example, patients with unfavorable changes in tumor function might be switched to a different chemotherapy and/or targeted regiment to achieve a better outcome. In contrast, patients with favorable changes in tumor features may go on receiving the original therapy regimen and rely on more conservative strategies in their future course. In this prospective study, we used the molecular tracer 99m Tc-3PRGD2 to measure response. 99mTc-3PRGD2 is a molecular imaging probe that has a positive correlation with the expression integrin avb3, because it utilizes a refined dimeric RGD peptide that has an enhanced binding affinity and improved tumor uptake as shown in preclinical
Ann Nucl Med
experiments [20–23]. It is generally regarded that the T/N ratio of RGD-related imaging can reflect angiogenesis in tumors. Therefore, 99mTc-3PRGD2 imaging can be used to predict treatment response indirectly by evaluating the angiogenesis status. By monitoring 28 advanced nonsquamous NSCLC patients, we demonstrated that 99mTc-3PRGD2 SPECT/CT during treatment can achieve a favorable predictive outcome. No significant difference in T/N ratio of SPECT1 between responders and nonresponders was observed. This means that the pre-therapy uptake was not predictive of response. However, T/N ratio decreased more in responders compared with nonresponders at SPECT2 and SPECT3 (p \ 0.05). Additionally, the responders showed a larger initial drop in T/N ratio at SPECT2, whereas the nonresponders showed minimal change as shown in Fig. 2. There is an inverse correlation between T/N ratio and response to chemoradiotherapy plus bevacizumab in responders. That is to say, a declining T/N ratio and rising T/N ratio after treatment often herald a good and poor effect, respectively. These findings suggest that changes in tumor function may be more suitable for predicting parameters. There were three outliers in our study. The first outlier showed a poor response at the second time point, but achieved a complete response at completion of therapy. We speculated at one point whether there had been some postradiation changes or the patient had had an infectious or inflammatory process in this area. But the patients had no related symptoms and other examination results did not support this either and so it was ultimately discarded. After discussion with oncologists, we think this result may be due to the heterogeneous nature of the tumor. Various reports have demonstrated that the tumor microenvironment can be highly heterogeneous [24]. The microenvironment in some parts of a solid tumor can be hypoxic and poorly supplied with nutrients because of inadequate blood vessel networks and an imbalance between proliferation and angiogenesis [25, 26]. Therefore, tumor cells can show quite different characteristics of cell activity, including proliferation and dormancy [27, 28]. Tumor cells in a hypoxic region distant from blood vessels show decreased proliferation or dormancy and resistance to chemo- or radiotherapy, which occurs occasionally during tumor treatment [29–32]. Whereas integrin avb3 is overexpressed on both tumor cells and sprouting tumor vasculature, it is not expressed on most normal tissues or resting endothelial cells. There is a positive correlation between integrin avb3 and 99m Tc-3P RGD2 uptake, the peptide used in our tracer [12]. According to our analysis with oncologists, some tumor cells of this patient may have been in the dormancy stage during the first half of the treatment period and converted to a proliferating stage during the second half. This may be the reason for the relatively low initial T/N ratio of this
tumor and little degree of change from SPECT1 to SPECT2. The second PR outlier had an initial increase of T/N ratio from SPECT1 to SPECT2 with a subsequent reduction in T/N ratio from SPECT1 to SPECT3. This patient had a transient fever during treatment, and the lesion image seemed to become larger with a fuzzy edge. The patient was diagnosed with lung cancer-associated inflammation. After a period of anti-inflammatory treatment, the patient’s temperature returned to normal, and the lesion image became smaller with a clear edge. Papetti et al. [33] reported that angiogenesis occurs in certain physiological processes, such as in infections as well as tumors in adults, causing an increase in the number of integrin avb3. This may have caused the T/N ratio of this tumor to increase temporarily in the second outlier case. The third outlier was a PD patient with a marked reduction from SPECT1 to SPECT3 (57.3 % reduction in uptake). But the lesion became larger with necrotic tissue in its center, which was identified by the final CT. Because there was almost no angiogenesis and only relatively incomplete blood vessel networks in the necrotic tissue, the tracer could not enter this area and so the average T/N ratio fell to a low level. Considering these outliers, attempting to utilize a single time point during the treatment course to predict response may not be ideal. Multiple time point imaging can offer more accurate information for the prediction of therapy over a single time point to enable better management of the disease. In our study, semiquantitative analysis of the change in 99m Tc-3P RGD2 uptake was successful in identifying responders to treatment. A reduction of greater than 25.9 % in T/N ratio can differentiate responders from nonresponders as early as 3–5 weeks after the initiation of chemoradiotherapy plus bevacizumab. The AUC for determining the therapeutic effect using T/N ratios change at SPECT2 and SPECT3 was 0.856 and 0.909, respectively. The value of SPECT2 is consistent with the value reported by Huang et al. using FDG-PET at the same time point [18]. Yet, we had to admit the existence of an overlap between responders and nonresponders due to infection or necrosis. It was not determined which treatment method played the main role and whether the change of angiogenesis benefited from chemoradiotherapy, bevacizumab or some synergy between them. These shortcomings cannot be resolved though our existing technique and is a limitation of this study. In the follow-up of our study, we found that the functional response was significantly associated with clinical outcome. The one-year cumulative survival rates for P1 responders, P1 nonresponders, P2 responders and P2 nonresponders were 91.7, 56.3, 82.4 and 63.6 %, respectively.
123
Ann Nucl Med
There is a significant difference in OS between P1 responders and P1 nonresponders (p \ 0.05), and also between P2 responders and P2 nonresponders (p \ 0.05). Persistently high uptake by the tumor was always associated with a poor prognosis. Currently, there is a functional response classification system like PRECIST, but it is not well suited for SPECT and SPECT/CT imaging. There is a need to standardize SPECT response classification by enlarging the sample number, which will be the focus of our future work. In conclusion, we demonstrated the ability to differentiate tumor functional response during the early stages of therapy by quantitative assessment of 99mTc-3PRGD2 uptake in the current study. SPECT/CT after two cycles of chemoradiotherapy plus bevacizumab can predict patients who will have a better response to treatment and survival. Acknowledgments This research was supported by the National Natural Science Foundation of China (NSFC) projects (81271606), Research Fund of Science and Technology Department of Jilin Province (20150520154JH).
References 1. International Agency for Research on Cancer. Lung cancer estimated incidence. Mortality and Prevalence Worldwide in 2012. 2. Jemal A, Siegel R, Xu J, Ward E. Cancer statistics, 2010. CA Cancer J Clin. 2010;60(5):277–300. 3. William WN Jr, Lin HY, Lee JJ, Lippman SM, Roth JA, Kim ES. Revisiting stage IIIB and IV non-small cell lung cancer: analysis of the surveillance, epidemiology, and end results data. Chest. 2009;136(3):701–9. 4. Zhang GB, Chen J, Wang LR, Li J, Li MW, Xu N, et al. RRM1 and ERCC1 expression in peripheral blood versus tumor tissue in gemcitabine/carboplatin-treated advanced non-small cell lung cancer. Cancer Chemother Pharmacol. 2012;69(5):1277–87. 5. Sun S, Schiller JH, Spinola M, Minna JD. New molecularly targeted therapies for lung cancer. J Clin Invest. 2007;117(10): 2740–50. 6. Nuijten MJ, Aultman R, Carpeno Jde C, Vergnenegre A, Chouaid C, Walzer S, et al. An indirect comparison of the efficacy of bevacizumab plus carboplatin and paclitaxel versus pemetrexed with cisplatin in patients with advanced or recurrent non-squamous adenocarcinoma non-small cell lung cancer. Curr Med Res Opin. 2011;27(11):2193–201. 7. Dinglin XX, Huang Y, Liu H, Zeng YD, Hou X, Chen LK. Pemetrexed and cisplatin combination with concurrent whole brain radiotherapy in patients with brain metastases of lung adenocarcinoma: a single-arm phase II clinical trial. J Neurooncol. 2013;112(3):461–6. 8. van Gool MH, Aukema TS, Hartemink KJ, Valdes Olmos RA, van Tinteren H, Klomp HM. FDG-PET/CT response evaluation during EGFR-TKI treatment in patients with NSCLC. World J Radiol. 2014;6(7):392–8. 9. van Gool MH, Aukema TS, Schaake EE, Rijna H, Codrington HE, Valdes Olmos RA, et al. (18)F-fluorodeoxyglucose positron emission tomography versus computed tomography in predicting histopathological response to epidermal growth factor receptortyrosine kinase inhibitor treatment in resectable non-small cell lung cancer. Ann Surg Oncol. 2014;21(9):2831–7.
123
10. Yang TJ, Aukema TS, van Tinteren H, Burgers S, Valdes Olmos R, Verheij M. Predicting early chemotherapy response with technetium-99m methoxyisobutylisonitrile SPECT/CT in advanced non-small cell lung cancer. Mol Imaging Biol. 2010; 12(2):174–80. 11. Niu G, Chen X. Why integrin as a primary target for imaging and therapy. Theranostics. 2011;1:30–47. 12. Chen X, Sievers E, Hou Y, Park R, Tohme M, Bart R, et al. Integrin alpha v beta 3-targeted imaging of lung cancer. Neoplasia. 2005;7(3):271–9. 13. Zhu Z, Miao W, Li Q, Dai H, Ma Q, Wang F, et al. 99mTc3PRGD2 for integrin receptor imaging of lung cancer: a multicenter study. J Nucl Med. 2012;53(5):716–22. 14. Ma Q, Ji B, Jia B, Gao S, Ji T, Wang X, et al. Differential diagnosis of solitary pulmonary nodules using (9)(9)mTc-3P(4)RGD(2) scintigraphy. Eur J Nucl Med Mol Imaging. 2011; 38(12):2145–52. 15. Ji S, Zhou Y, Voorbach MJ, Shao G, Zhang Y, Fox GB, et al. Monitoring tumor response to linifanib therapy with SPECT/CT using the integrin alphavbeta3-targeted radiotracer 99mTc-3PRGD2. J Pharmacol Exp Ther. 2013;346(2):251–8. 16. Ferrara N, Hillan KJ, Gerber HP, Novotny W. Discovery and development of bevacizumab, an anti-VEGF antibody for treating cancer. Nat Rev Drug Discov. 2004;3(5):391–400. 17. Dudek AZ, Mahaseth H. Circulating angiogenic cytokines in patients with advanced non-small cell lung cancer: correlation with treatment response and survival. Cancer Invest. 2005;23(3):193–200. 18. Huang W, Zhou T, Ma L, Sun H, Gong H, Wang J, et al. Standard uptake value and metabolic tumor volume of (1)(8)F-FDG PET/ CT predict short-term outcome early in the course of chemoradiotherapy in advanced non-small cell lung cancer. Eur J Nucl Med Mol Imaging. 2011;38(9):1628–35. 19. Abdallah A, Belal M, El Bastawisy A, Gaafar R. Plasma vascular endothelial growth factor 165 in advanced non-small cell lung cancer. Oncol Lett. 2014;7(6):2121–9. 20. Haubner R, Wester HJ, Weber WA, Mang C, Ziegler SI, Goodman SL, et al. Noninvasive imaging of alpha(v)beta3 integrin expression using 18F-labeled RGD-containing glycopeptide and positron emission tomography. Cancer Res. 2001;61(5):1781–5. 21. Beer AJ, Haubner R, Goebel M, Luderschmidt S, Spilker ME, Wester HJ, et al. Biodistribution and pharmacokinetics of the alphavbeta3-selective tracer 18F-galacto-RGD in cancer patients. J Nucl Med. 2005;46(8):1333–41. 22. Kenny LM, Coombes RC, Oulie I, Contractor KB, Miller M, Spinks TJ, et al. Phase I trial of the positron-emitting Arg-GlyAsp (RGD) peptide radioligand 18F-AH111585 in breast cancer patients. J Nucl Med. 2008;49(6):879–86. 23. Bach-Gansmo T, Danielsson R, Saracco A, Wilczek B, Bogsrud TV, Fangberget A, et al. Integrin receptor imaging of breast cancer: a proof-of-concept study to evaluate 99mTc-NC100692. J Nucl Med. 2006;47(9):1434–9. 24. Mardilovich K, Shaw LM. Hypoxia regulates insulin receptor substrate-2 expression to promote breast carcinoma cell survival and invasion. Cancer Res. 2009;69(23):8894–901. 25. Pouyssegur J, Dayan F, Mazure NM. Hypoxia signalling in cancer and approaches to enforce tumour regression. Nature. 2006;441(7092):437–43. 26. Denko NC. Hypoxia, HIF1 and glucose metabolism in the solid tumour. Nat Rev Cancer. 2008;8(9):705–13. 27. Okuyama H, Endo H, Akashika T, Kato K, Inoue M. Downregulation of c-MYC protein levels contributes to cancer cell survival under dual deficiency of oxygen and glucose. Cancer Res. 2010;70(24):10213–23. 28. Tannock IF. The relation between cell proliferation and the vascular system in a transplanted mouse mammary tumour. Br J Cancer. 1968;22(2):258–73.
Ann Nucl Med 29. Tredan O, Galmarini CM, Patel K, Tannock IF. Drug resistance and the solid tumor microenvironment. J Natl Cancer Inst. 2007;99(19):1441–54. 30. Tannock IF. Tumor physiology and drug resistance. Cancer Metastasis Rev. 2001;20(1–2):123–32. 31. Saggar JK, Yu M, Tan Q, Tannock IF. The tumor microenvironment and strategies to improve drug distribution. Front Oncol. 2013;3:154.
32. Endo H, Okuyama H, Ohue M, Inoue M. Dormancy of cancer cells with suppression of AKT activity contributes to survival in chronic hypoxia. PLoS ONE. 2014;9(6):e98858. 33. Papetti M, Herman IM. Mechanisms of normal and tumorderived angiogenesis. Am J Physiol Cell Physiol. 2002;282(5): C947–70.
123
ORIGINAL ARTICLE
99m
Tc-3PRGD2 SPECT/CT predicts the outcome of advanced nonsquamous non-small cell lung cancer receiving chemoradiotherapy plus bevacizumab
Qingjie Ma1 • Kaiyin Min1 • Ting Wang1 • Bin Chen1 • Qiang Wen1 Fan Wang2 • Tiefeng Ji1 • Shi Gao1
•
Received: 11 March 2015 / Accepted: 12 April 2015 Ó The Japanese Society of Nuclear Medicine 2015
Abstract Background Functional imaging can help clinicians assess the individual response of advanced nonsquamous non-small cell lung cancer (NSCLC) to chemoradiation therapy plus bevacizumab. Our purpose is to investigate the ability of 99mTc-3PRGD2 single photon emission computed tomography/computed tomography (SPECT/CT) in predicting the early response to treatment. Methods Patients with advanced nonsquamous NSCLC diagnosed by histological or cytological examination were imaged with 99mTc-3PRGD2 SPECT/CT at 3 time points: 1–3 days before the start of treatment (SPECT1), 40 Gy radiotherapy with 2 cycles of chemotherapy plus bevacizumab (SPECT2) and 4 weeks after chemoradiotherapy plus bevacizumab (SPECT3). The images were evaluated semiquantitatively by measuring the tumor to non-tumor ratio (T/N) and calculating the percentage change in T/N ratio. Short-term outcome was assessed by the treatment response evaluation according to the Response Evaluation Criteria in Solid Tumors criteria as: complete response (CR), partial response (PR), stable disease (SD) and progressive disease (PD). Patients were divided two groups: responders (CR and PR) and nonresponders (SD and PD). To determine a threshold for percent reduction in T/N ratios, receiver-operating characteristic (ROC) curve analysis & Tiefeng Ji [email protected] & Shi Gao [email protected] 1
Department of Nuclear Medicine, China-Japan Union Hospital of Jilin University, Changchun, China
2
Medical Isotopes Research Center, Peking University, Beijing, China
was used. Patients were grouped again based on the threshold of P1 (the change percentage from SPECT1 to SPECT2) and P2 (the change percentage from SPECT1 to SPECT3): P1 responders and P1 nonresponders; P2 responders and P2 nonresponders. Patients were followed up starting 4 weeks after completion of therapy and then every 3 months for the first 2 years and every 6 months after 2 years. OS of P1 responders, P1 nonresponders, P2 responders and P2 nonresponders was estimated and graphically illustrated using the Kaplan–Meier method and the log-rank test was used to test the null hypotheses of equal OS in subgroups of patients. Results A total of 28 patients completed all imaging and treatment. All primary lung tumors were well visualized on SPECT1. The mean T/N ratio of SPECT1 in responders and nonresponders was not statistically different (2.73 ± 0.59 vs. 2.59 ± 0.52, p [ 0.05). At SPECT2 and SPECT3, the mean T/N ratios were both lower in the responders compared with the nonresponders and had statistical significance (p \ 0.05). P1 and P2 in the responders was larger than the nonresponders with significant difference (P1: 34.18 ± 21.55 % vs. 9.02 ± 14.02 %, p \ 0.05; P2: 53.02 ± 15.50 % vs. 7.74 ± 37.95 %, p \ 0.05). The optimal threshold of P1 that can discriminate between P1 responders and P1 nonresponders was greater than 25.9 % reduction, and that of P2 that can discriminate between P2 responders and P2 nonresponders was 34.0 % reduction. The area under the ROC curve (AUC) of P1 and P2 for determining residual disease was 0.856 and 0.909, respectively; but there was no statistical significance between them (p [ 0.05). There was a significant difference for OS between P1 responders and P1 nonresponders (p \ 0.05), and also for OS between P2 responders and P2 nonresponders (p \ 0.05). But there was no difference between the P1 responders and P2 responders (p [ 0.05), or
123
Ann Nucl Med
between the P1 nonresponders and P2 nonresponders (p [ 0.05). Conclusion A 99mTc-3PRGD2 SPECT/CT after two cycles of chemoradiotherapy plus bevacizumab can predict patients who will have a better response to treatment and survival. Keywords 99mTc-3PRGD2 SPECT/CT Lung cancer Response Prediction Bevacizumab
by histological or cytological examination were enrolled in this study at China–Japan Union Hospital, Changchun, China. These patients are not indicated for surgery due to metastases or other co-morbidities. The tumors were measurable with positive imaging according to Response Evaluation Criteria in Solid Tumors (RECIST). Informed written consent to participate in this study was obtained from all patients. The study protocol was approved by the ethics committee and the Institutional Review Board of China–Japan Union Hospital, Changchun, China.
Introduction
Treatment
Lung cancer ranks as one of the most common types of cancer in China and worldwide with an increasing incidence rate. It is the leading cause of malignant disease mortality [1, 2]. Nearly 80 % of lung cancer cases are nonsmall cell lung cancer (NSCLC), and the overall five-year survival of such cases can barely reach 15 % [3–5]. Although great advances have been achieved in diagnostics and treatment, metastatic disease still occurs in 70 % of NSCLC patients after receiving treatment [6, 7]. Therefore, accurate methods that enable outcome prediction to screen patients who are likely to benefit from treatment are required urgently. Functional imaging, such as positron emission tomography/computed tomography (PET/CT) and single photon emission computed tomography/CT (SPECT/CT), can both help clinicians design individualized treatment to improve the outcome of advanced NSCLC [8–10]. Novel avb3specific tracers for SPECT, like technetium-99m (99mTc) radiolabeled Arg-Gly-Asp (RGD) peptides and analogs, can specifically target the integrin avb3, which plays a critical role in the regulation of tumor angiogenesis and metastasis [11, 12]. Therefore, such tracers have been intensively investigated for noninvasive functional imaging of tumors in pre-clinical and clinical studies. 99mTc3PRGD2 SPECT/CT imaging can provide more accurate and timelier information than structural imaging alone [13– 15]. Therefore, we conducted this prospective study to investigate the correlation between different tumor angiogenesis changes with the end-of-treatment response to determine whether SPECT can predict short-term outcomes of chemoradiotherapy plus bevacizumab in patients with advanced nonsquamous NSCLC.
Patients received concomitant chemoradiotherapy plus bevacizumab designed by medical oncologists according to international standards. The chemotherapy regimen used in this study was cisplatin (administered i.v. at 80 mg/m2 on day 1) and gemcitabine (i.v. at 1250 mg/m2 on days 1 and 8), repeated every 3 weeks for up to six cycles. Bevacizumab was administered i.v. at 15 mg/kg concurrently with chemotherapy every 3 weeks on day 1. The radiotherapy regimen was intensity-modulated radiotherapy (IMRT). The patients were treated with a conventional fractionation scheme 5 days per week with a fractional dose of 5 9 2 Gy per week and late course accelerated hyperfractionated RT with 1.4 Gy twice daily up to a total cumulative dose of 63–67.5 Gy. RT was based on a planning CT image. The gross tumor volume (GTV) was delineated to the primary tumor and metastatic regional lymph nodes, and the planning target volume included the GTV with a margin of 1.0–1.5 cm.
Materials and methods Patients From May 2011 to January 2014,patients with nonsquamous advanced NSCLC (IIIB or IV tumor stage) diagnosed
123
99m
99m
Tc-3PRGD2 SPECT/CT imaging
Tc-3PRGD2 SPECT/CT imaging was performed at three time points: SPECT1 (1–3 days before the start of treatment), SPECT2 (40 Gy IMRT with 2 cycles of chemotherapy plus bevacizumab) and SPECT3 (4 weeks after completion of treatment) using a SPECT/CT scanner (Philips Precedence). Radiolabeling and quality control procedures for 3PRGD2 were performed as described previously [14]. 3PRGD2 was radiolabeled with 378 ± 86 MBq 99m technetium and thereafter administered via a single intravenous bolus injection in the contralateral arm to the affected lung, followed by a 10 mL saline flush. During imaging, patients were in the supine position with raised arms. X-ray scanning by a double-head c camera was performed to show the detecting range. The following CT scanning was set at a matrix of 256 9 256 and a 5-mm layer thickness. After collecting CT images, the detecting bench was positioned automatically for
Ann Nucl Med
SPECT data collection. The matrix was 128 9 128 pixels, and the photopeak was centered at 140 keV with a symmetrical 20 % window. Imaging with both radiotracers was performed using 6° angular steps in a 20 s time frame. The distance between the chest and detector was minimized. The Digital Imaging and Communications in Medicine image files of each patient were saved on optical disks and transferred to Extended BrillianceTM workspace (Philips Healthcare) for centralized reconstruction, reading and analysis. Imaging analysis All images were visually analyzed by two experienced nuclear medicine physicians without knowledge of the patients’ history. For semiquantitative analysis of tracer uptake in tumors following SPECT1, SPECT2 and SPECT3, regions of interest (ROIs) were drawn around the entire tumor and to the contralateral normal lung tissue with the same size of the tumor, and tumor to non-tumor localization (T/N) ratios of 99mTc-3PRGD2 SPECT/CT were determined. We used T/N ratios for further quantitative assessment and potential differentiation of lesions. The percentage changes in T/N ratios between SPECT1 and SPECT2 (P1) or SPECT3 (P2) were calculated using the following formulae: P1 ¼ ½T=NSPECT1 T=NSPECT2 =T=NSPECT1 100 %; P2 ¼ ½T=NSPECT1 T=NSPECT3 =T=NSPECT1 100 %: Clinical response validation Clinical short-term outcome was verified using CT of SPECT3 according to RECIST criteria as: complete response (CR), partial response (PR), stable disease (SD) and progressive disease (PD). Patients were divided into two groups: responders (CR and PR) and nonresponders (SD and PD). Patient follow-up Patients were followed up starting at 4 weeks after completion of therapy and then every 3 months for the first 2 years and every 6 months after 2 years. The follow-up includes history taking and physical examination. Overall survival (OS) was defined as the time from the completion of treatment to death from any cause. Statistical analysis All data were expressed as the mean ± standard deviation (SD). Statistical analyses were performed using software
SPSS for Windows (version 19.0). The differences in the dosimetric parameters between responders and nonresponders were evaluated by t test. To determine a threshold for percent reduction in T/N ratios, receiver-operating characteristic (ROC) curve analysis was used. The area under the ROC curve (AUC) provided the predictive power for 99mTc-3PRGD2 SPECT/CT imaging. Patients were grouped again based on the threshold of P1 and P2: P1 responders and P1 nonresponders; P2 responders and P2 nonresponders. OS of P1 responders, P1 nonresponders, P2 responders and P2 nonresponders was estimated and graphically illustrated using the Kaplan–Meier method, and the log-rank test was used to test the null hypotheses of equal OS in subgroups of patients. p \ 0.05 was considered to indicate statistical significance.
Results Patients Our original plan was to recruit thirty patients for this study, but one patient voluntarily dropped out and one patient lost contact. These two cases were excluded from further analysis. The remaining 28 patients completed all 3 SPECT/CT phases and treatment. Patient characteristics are shown in Table 1. The mean age was 55.1 ± 13.2 years (range 32–67 years). The mean followup time for all patients was 15 months (3–30 months).
Table 1 Characteristics of 28 patients enrolled in the study Characteristic
Number
Sex Male
12
Female
16
Mean age ± SD
55.1 ± 13.2
Tumor size, pretreatment (CT), mm ± SD
58.5 ± 14.7
Clinical stage IIIB
22
IV
6
Pathological type Adenocarcinoma
24
Large cell carcinoma Mucoepidermoid carcinoma
3 1
Clinical response Complete response
5
Partial response
12
Stable disease
8
Progressive disease
3
123
Ann Nucl Med
Clinical response Based on RECIST, 5 patients achieved CR and 12 had PR. The overall response rate was 60.71 %. By comparison, 8 patients attained SD and only 3 patients had PD. T/N ratios All primary lung tumors were well visualized on SPECT1 and evaluated using T/N ratios. The absolute T/N ratios and change in T/N ratios of three time points determined by two observers are summarized in Table 2. The mean T/N ratio of SPECT1 in all patients was 2.68 ± 0.56. No significant difference was observed in the mean pre-therapy T/N ratio between responders and nonresponders (2.73 ± 0.59 vs. 2.59 ± 0.52, p [ 0.05). At SPECT2 and SPECT3, the mean T/N ratios were both lower in the responders compared with the nonresponders and had statistical significance (SPECT2: 1.72 ± 0.46 vs. 2.33 ± 0.46, p \ 0.05; SPECT3: 1.23 ± 0.31 vs. 2.30 ± 0.74, p \ 0.05). P1 in the responders was larger than the nonresponders with significant difference (34.18 ± 21.55 vs. 9.02 ± 14.02, p \ 0.05). The P2 between the two groups also had a significant difference (53.02 ± 15.50 vs. 7.74 ± 37.95, p \ 0.05). Figure 1 shows typical examples of 99mTc-3PRGD2 SPECT/CT in patients with responding and nonresponding tumors. Figure 2 shows the different patterns of changes in T/N ratios of the responders and nonresponders at three time points. The responders showed a larger initial drop in T/N ratio at SPECT2 from SPECT1, whereas the nonresponders showed a minimal drop in the ratio and even higher values. The three outliers in Fig. 2 demonstrated an unexpected change in the pattern of uptake between SPECT2 and SPECT3. The first outlier was in the CR group. During SPECT2, the outlier’s T/N ratio demonstrated a poor response to treatment (10.6 % reduction in uptake), but subsequently showed a 51.4 % reduction in T/N ratio at SPECT3 with no visual image on the final SPECT image. The second outlier was a PR patient. There was an early 5.6 % increase in the T/N ratio from SPECT1 to SPECT2, but a 12.2 % reduction in T/N ratio from SPECT1 to SPECT3. In the final CT, the patient was found to have
Table 2 T/N ratios and its changes in three time points of SPECT/CT
T/N ratio
All patients
mild residual disease. The third outlier was a PD patient with a marked reduction from SPECT1 to SPECT3 (57.3 % reduction in uptake). But the lesion became larger, as identified by the final CT. Figure 3 illustrates P1 and P2 according to RECIST groupings. As seen in Fig. 3, the thresholds at P1 and P2 were 25.9 and 34.0 % drop change. ROC curve analysis The optimal threshold of P1 that can discriminate between P1 responders and P1 nonresponders was greater than 25.9 % reduction, and the optimal threshold of P2 that can discriminate between P2 responders and P2 nonresponders was 34.0 % reduction. The AUC of the ROCs for P1 and P2 was 0.856 and 0.909, respectively (Fig. 4). The predicting ability of P2 was higher than P1, but without statistical significance (p [ 0.05). Tumor response was a significant predictor of OS in the analysis. The one-year cumulative survival rates for P1 responders, P1 nonresponders, P2 responders and P2 nonresponders were 91.7, 56.3, 82.4 and 63.6 %. There was a significant difference for OS between P1 responders and P1 nonresponders (p \ 0.05), and also for OS between P2 responders and P2 nonresponders (p \ 0.05; Fig. 5). But there was no difference between the P1 responders and P2 responders (p [ 0.05), or between the P1 nonresponders and P2 nonresponders (p [ 0.05).
Discussion Patients with advanced NSCLC have a higher risk of cancer-related death in comparison with less advanced stages of the cancer. Concomitant chemoradiotherapy is widely used for first-line treatment of advanced NSCLC. Bevacizumab is an anti-vascular endothelial growth factor mAb with good clinical benefits established in a variety of tumors, including lung cancer [16]. So in nonsquamous NSCLC, bevacizumab is always used with concomitant chemoradiotherapy on the basis of its favorable efficacy and tolerability profile. However, the response to this therapeutic regimen is variable [17, 18]. Previous studies
Responder group*
Nonresponder group*
SPECT1
2.68 ± 0.56
2.73 ± 0.59
2.59 ± 0.52
0.527
SPECT2
1.96 ± 0.55
1.72 ± 0.46
2.33 ± 0.46
0.002
SPECT3
1.65 ± 0.74
1.23 ± 0.31
2.30 ± 0.74
0.001
P1 (%)
24.30 ± 22.46
34.18 ± 21.55
9.02 ± 14.02
0.001
P2 (%)
35.23 ± 34.39
53.02 ± 15.50
7.74 ± 37.95
0.003
p* is the comparative value of Responder group and Nonresponder group
123
p*
Ann Nucl Med
Fig. 1 Two typical examples of 99mTc-3PRGD2 SPECT/CT images in responders (a) and nonresponders (b) acquired at SPECT1, SPECT2 and SPECT3. For the responder, the mass size decreased significantly and the mass was almost invisible during the final
SPECT/CT (red arrows). For the nonresponder, necrosis appeared at the center of the lesion (blue arrows), and pathologic fractures from bone metastases (pink arrow) appeared at the same time (color figure online)
Fig. 2 The changes in the T/N ratio of the two groups (4 subgroups) at the three time points
123
Ann Nucl Med
Fig. 3 Percent changes in the T/N ratio of
99m
Tc-3PRGD2 of P1 (a) and P2 (b) of the responders versus nonresponders
Fig. 4 Comparison between ROC analysis of P1 and P2
have shown that patients achieving a good response to therapeutic regimens have a better prognosis than those with poor response [18, 19]. Long-term ineffective treatment not only brings great pain to patients, but also increases their economic burden. Therefore, there is an urgent need to evaluate the effect of therapy in a timely manner. Unfortunately, the limitations of conventional tools (X-ray, CT) to measure response make it difficult to accurately assess the quantity of residual viable tumor over the course of treatment [18]. Molecular imaging can provide information about the changes of the tumor on the molecular level and quantify the amount of viable residual disease over the course of treatment. Based on the response to therapy, clinicians can make individualized treatment
123
Fig. 5 OS of P1 responders, P2 responders, P1 nonresponders and P2 nonresponders was estimated by the Kaplan–Meier method
adjustments quickly. For example, patients with unfavorable changes in tumor function might be switched to a different chemotherapy and/or targeted regiment to achieve a better outcome. In contrast, patients with favorable changes in tumor features may go on receiving the original therapy regimen and rely on more conservative strategies in their future course. In this prospective study, we used the molecular tracer 99m Tc-3PRGD2 to measure response. 99mTc-3PRGD2 is a molecular imaging probe that has a positive correlation with the expression integrin avb3, because it utilizes a refined dimeric RGD peptide that has an enhanced binding affinity and improved tumor uptake as shown in preclinical
Ann Nucl Med
experiments [20–23]. It is generally regarded that the T/N ratio of RGD-related imaging can reflect angiogenesis in tumors. Therefore, 99mTc-3PRGD2 imaging can be used to predict treatment response indirectly by evaluating the angiogenesis status. By monitoring 28 advanced nonsquamous NSCLC patients, we demonstrated that 99mTc-3PRGD2 SPECT/CT during treatment can achieve a favorable predictive outcome. No significant difference in T/N ratio of SPECT1 between responders and nonresponders was observed. This means that the pre-therapy uptake was not predictive of response. However, T/N ratio decreased more in responders compared with nonresponders at SPECT2 and SPECT3 (p \ 0.05). Additionally, the responders showed a larger initial drop in T/N ratio at SPECT2, whereas the nonresponders showed minimal change as shown in Fig. 2. There is an inverse correlation between T/N ratio and response to chemoradiotherapy plus bevacizumab in responders. That is to say, a declining T/N ratio and rising T/N ratio after treatment often herald a good and poor effect, respectively. These findings suggest that changes in tumor function may be more suitable for predicting parameters. There were three outliers in our study. The first outlier showed a poor response at the second time point, but achieved a complete response at completion of therapy. We speculated at one point whether there had been some postradiation changes or the patient had had an infectious or inflammatory process in this area. But the patients had no related symptoms and other examination results did not support this either and so it was ultimately discarded. After discussion with oncologists, we think this result may be due to the heterogeneous nature of the tumor. Various reports have demonstrated that the tumor microenvironment can be highly heterogeneous [24]. The microenvironment in some parts of a solid tumor can be hypoxic and poorly supplied with nutrients because of inadequate blood vessel networks and an imbalance between proliferation and angiogenesis [25, 26]. Therefore, tumor cells can show quite different characteristics of cell activity, including proliferation and dormancy [27, 28]. Tumor cells in a hypoxic region distant from blood vessels show decreased proliferation or dormancy and resistance to chemo- or radiotherapy, which occurs occasionally during tumor treatment [29–32]. Whereas integrin avb3 is overexpressed on both tumor cells and sprouting tumor vasculature, it is not expressed on most normal tissues or resting endothelial cells. There is a positive correlation between integrin avb3 and 99m Tc-3P RGD2 uptake, the peptide used in our tracer [12]. According to our analysis with oncologists, some tumor cells of this patient may have been in the dormancy stage during the first half of the treatment period and converted to a proliferating stage during the second half. This may be the reason for the relatively low initial T/N ratio of this
tumor and little degree of change from SPECT1 to SPECT2. The second PR outlier had an initial increase of T/N ratio from SPECT1 to SPECT2 with a subsequent reduction in T/N ratio from SPECT1 to SPECT3. This patient had a transient fever during treatment, and the lesion image seemed to become larger with a fuzzy edge. The patient was diagnosed with lung cancer-associated inflammation. After a period of anti-inflammatory treatment, the patient’s temperature returned to normal, and the lesion image became smaller with a clear edge. Papetti et al. [33] reported that angiogenesis occurs in certain physiological processes, such as in infections as well as tumors in adults, causing an increase in the number of integrin avb3. This may have caused the T/N ratio of this tumor to increase temporarily in the second outlier case. The third outlier was a PD patient with a marked reduction from SPECT1 to SPECT3 (57.3 % reduction in uptake). But the lesion became larger with necrotic tissue in its center, which was identified by the final CT. Because there was almost no angiogenesis and only relatively incomplete blood vessel networks in the necrotic tissue, the tracer could not enter this area and so the average T/N ratio fell to a low level. Considering these outliers, attempting to utilize a single time point during the treatment course to predict response may not be ideal. Multiple time point imaging can offer more accurate information for the prediction of therapy over a single time point to enable better management of the disease. In our study, semiquantitative analysis of the change in 99m Tc-3P RGD2 uptake was successful in identifying responders to treatment. A reduction of greater than 25.9 % in T/N ratio can differentiate responders from nonresponders as early as 3–5 weeks after the initiation of chemoradiotherapy plus bevacizumab. The AUC for determining the therapeutic effect using T/N ratios change at SPECT2 and SPECT3 was 0.856 and 0.909, respectively. The value of SPECT2 is consistent with the value reported by Huang et al. using FDG-PET at the same time point [18]. Yet, we had to admit the existence of an overlap between responders and nonresponders due to infection or necrosis. It was not determined which treatment method played the main role and whether the change of angiogenesis benefited from chemoradiotherapy, bevacizumab or some synergy between them. These shortcomings cannot be resolved though our existing technique and is a limitation of this study. In the follow-up of our study, we found that the functional response was significantly associated with clinical outcome. The one-year cumulative survival rates for P1 responders, P1 nonresponders, P2 responders and P2 nonresponders were 91.7, 56.3, 82.4 and 63.6 %, respectively.
123
Ann Nucl Med
There is a significant difference in OS between P1 responders and P1 nonresponders (p \ 0.05), and also between P2 responders and P2 nonresponders (p \ 0.05). Persistently high uptake by the tumor was always associated with a poor prognosis. Currently, there is a functional response classification system like PRECIST, but it is not well suited for SPECT and SPECT/CT imaging. There is a need to standardize SPECT response classification by enlarging the sample number, which will be the focus of our future work. In conclusion, we demonstrated the ability to differentiate tumor functional response during the early stages of therapy by quantitative assessment of 99mTc-3PRGD2 uptake in the current study. SPECT/CT after two cycles of chemoradiotherapy plus bevacizumab can predict patients who will have a better response to treatment and survival. Acknowledgments This research was supported by the National Natural Science Foundation of China (NSFC) projects (81271606), Research Fund of Science and Technology Department of Jilin Province (20150520154JH).
References 1. International Agency for Research on Cancer. Lung cancer estimated incidence. Mortality and Prevalence Worldwide in 2012. 2. Jemal A, Siegel R, Xu J, Ward E. Cancer statistics, 2010. CA Cancer J Clin. 2010;60(5):277–300. 3. William WN Jr, Lin HY, Lee JJ, Lippman SM, Roth JA, Kim ES. Revisiting stage IIIB and IV non-small cell lung cancer: analysis of the surveillance, epidemiology, and end results data. Chest. 2009;136(3):701–9. 4. Zhang GB, Chen J, Wang LR, Li J, Li MW, Xu N, et al. RRM1 and ERCC1 expression in peripheral blood versus tumor tissue in gemcitabine/carboplatin-treated advanced non-small cell lung cancer. Cancer Chemother Pharmacol. 2012;69(5):1277–87. 5. Sun S, Schiller JH, Spinola M, Minna JD. New molecularly targeted therapies for lung cancer. J Clin Invest. 2007;117(10): 2740–50. 6. Nuijten MJ, Aultman R, Carpeno Jde C, Vergnenegre A, Chouaid C, Walzer S, et al. An indirect comparison of the efficacy of bevacizumab plus carboplatin and paclitaxel versus pemetrexed with cisplatin in patients with advanced or recurrent non-squamous adenocarcinoma non-small cell lung cancer. Curr Med Res Opin. 2011;27(11):2193–201. 7. Dinglin XX, Huang Y, Liu H, Zeng YD, Hou X, Chen LK. Pemetrexed and cisplatin combination with concurrent whole brain radiotherapy in patients with brain metastases of lung adenocarcinoma: a single-arm phase II clinical trial. J Neurooncol. 2013;112(3):461–6. 8. van Gool MH, Aukema TS, Hartemink KJ, Valdes Olmos RA, van Tinteren H, Klomp HM. FDG-PET/CT response evaluation during EGFR-TKI treatment in patients with NSCLC. World J Radiol. 2014;6(7):392–8. 9. van Gool MH, Aukema TS, Schaake EE, Rijna H, Codrington HE, Valdes Olmos RA, et al. (18)F-fluorodeoxyglucose positron emission tomography versus computed tomography in predicting histopathological response to epidermal growth factor receptortyrosine kinase inhibitor treatment in resectable non-small cell lung cancer. Ann Surg Oncol. 2014;21(9):2831–7.
123
10. Yang TJ, Aukema TS, van Tinteren H, Burgers S, Valdes Olmos R, Verheij M. Predicting early chemotherapy response with technetium-99m methoxyisobutylisonitrile SPECT/CT in advanced non-small cell lung cancer. Mol Imaging Biol. 2010; 12(2):174–80. 11. Niu G, Chen X. Why integrin as a primary target for imaging and therapy. Theranostics. 2011;1:30–47. 12. Chen X, Sievers E, Hou Y, Park R, Tohme M, Bart R, et al. Integrin alpha v beta 3-targeted imaging of lung cancer. Neoplasia. 2005;7(3):271–9. 13. Zhu Z, Miao W, Li Q, Dai H, Ma Q, Wang F, et al. 99mTc3PRGD2 for integrin receptor imaging of lung cancer: a multicenter study. J Nucl Med. 2012;53(5):716–22. 14. Ma Q, Ji B, Jia B, Gao S, Ji T, Wang X, et al. Differential diagnosis of solitary pulmonary nodules using (9)(9)mTc-3P(4)RGD(2) scintigraphy. Eur J Nucl Med Mol Imaging. 2011; 38(12):2145–52. 15. Ji S, Zhou Y, Voorbach MJ, Shao G, Zhang Y, Fox GB, et al. Monitoring tumor response to linifanib therapy with SPECT/CT using the integrin alphavbeta3-targeted radiotracer 99mTc-3PRGD2. J Pharmacol Exp Ther. 2013;346(2):251–8. 16. Ferrara N, Hillan KJ, Gerber HP, Novotny W. Discovery and development of bevacizumab, an anti-VEGF antibody for treating cancer. Nat Rev Drug Discov. 2004;3(5):391–400. 17. Dudek AZ, Mahaseth H. Circulating angiogenic cytokines in patients with advanced non-small cell lung cancer: correlation with treatment response and survival. Cancer Invest. 2005;23(3):193–200. 18. Huang W, Zhou T, Ma L, Sun H, Gong H, Wang J, et al. Standard uptake value and metabolic tumor volume of (1)(8)F-FDG PET/ CT predict short-term outcome early in the course of chemoradiotherapy in advanced non-small cell lung cancer. Eur J Nucl Med Mol Imaging. 2011;38(9):1628–35. 19. Abdallah A, Belal M, El Bastawisy A, Gaafar R. Plasma vascular endothelial growth factor 165 in advanced non-small cell lung cancer. Oncol Lett. 2014;7(6):2121–9. 20. Haubner R, Wester HJ, Weber WA, Mang C, Ziegler SI, Goodman SL, et al. Noninvasive imaging of alpha(v)beta3 integrin expression using 18F-labeled RGD-containing glycopeptide and positron emission tomography. Cancer Res. 2001;61(5):1781–5. 21. Beer AJ, Haubner R, Goebel M, Luderschmidt S, Spilker ME, Wester HJ, et al. Biodistribution and pharmacokinetics of the alphavbeta3-selective tracer 18F-galacto-RGD in cancer patients. J Nucl Med. 2005;46(8):1333–41. 22. Kenny LM, Coombes RC, Oulie I, Contractor KB, Miller M, Spinks TJ, et al. Phase I trial of the positron-emitting Arg-GlyAsp (RGD) peptide radioligand 18F-AH111585 in breast cancer patients. J Nucl Med. 2008;49(6):879–86. 23. Bach-Gansmo T, Danielsson R, Saracco A, Wilczek B, Bogsrud TV, Fangberget A, et al. Integrin receptor imaging of breast cancer: a proof-of-concept study to evaluate 99mTc-NC100692. J Nucl Med. 2006;47(9):1434–9. 24. Mardilovich K, Shaw LM. Hypoxia regulates insulin receptor substrate-2 expression to promote breast carcinoma cell survival and invasion. Cancer Res. 2009;69(23):8894–901. 25. Pouyssegur J, Dayan F, Mazure NM. Hypoxia signalling in cancer and approaches to enforce tumour regression. Nature. 2006;441(7092):437–43. 26. Denko NC. Hypoxia, HIF1 and glucose metabolism in the solid tumour. Nat Rev Cancer. 2008;8(9):705–13. 27. Okuyama H, Endo H, Akashika T, Kato K, Inoue M. Downregulation of c-MYC protein levels contributes to cancer cell survival under dual deficiency of oxygen and glucose. Cancer Res. 2010;70(24):10213–23. 28. Tannock IF. The relation between cell proliferation and the vascular system in a transplanted mouse mammary tumour. Br J Cancer. 1968;22(2):258–73.
Ann Nucl Med 29. Tredan O, Galmarini CM, Patel K, Tannock IF. Drug resistance and the solid tumor microenvironment. J Natl Cancer Inst. 2007;99(19):1441–54. 30. Tannock IF. Tumor physiology and drug resistance. Cancer Metastasis Rev. 2001;20(1–2):123–32. 31. Saggar JK, Yu M, Tan Q, Tannock IF. The tumor microenvironment and strategies to improve drug distribution. Front Oncol. 2013;3:154.
32. Endo H, Okuyama H, Ohue M, Inoue M. Dormancy of cancer cells with suppression of AKT activity contributes to survival in chronic hypoxia. PLoS ONE. 2014;9(6):e98858. 33. Papetti M, Herman IM. Mechanisms of normal and tumorderived angiogenesis. Am J Physiol Cell Physiol. 2002;282(5): C947–70.
123
