Optimization of combined temozolomide and peptide receptor radionuclide therapy (PRRT) in mice after multimodality molecular imaging studies

Background Successful treatments of patients with somatostatin receptor (SSTR)-overexpressing neuroendocrine tumours (NET) comprise somatostatin-analogue lutetium-177-labelled octreotate (177Lu-TATE) treatment, also referred to as peptide receptor radionuclide therapy (PRRT), and temozolomide (TMZ) treatment. Their combination might result in additive effects. Using MRI and SPECT/CT, we studied tumour characteristics and therapeutic responses after different (combined) administration schemes in a murine tumour model in order to identify the optimal treatment schedule for PRRT plus TMZ. Methods We performed molecular imaging studies in mice bearing SSTR-expressing H69 (humane small cell lung cancer) tumours after single intravenous (i.v.) administration of 30 MBq 177Lu-TATE or TMZ (oral 50 mg/kg daily for 14 days). Tumour perfusion was evaluated weekly by dynamic contrast-enhanced MRI (DCE-MRI), whereas tumour uptake of 111In-octreotide was quantified using SPECT/CT until day 39 after treatment. Based on these results, seven different 177Lu-octreotate and TMZ combination schemes were evaluated for therapy response, varying the order and time interval of the two therapies and compared with single treatments. Results PRRT and TMZ both resulted in tumour size reduction, accompanied by significant changes in MRI characteristics such as an enhanced tumour perfusion. Moreover, TMZ treatment also resulted in increased uptake of the SST analogue 111In-octreotide until day 13. In the subsequent therapy study, 90 % of animals receiving 177Lu-TATE at day 14 after TMZ treatment showed complete response, being the best anti-tumour results among groups. Conclusions Molecular imaging studies indicated that PRRT after TMZ treatment could induce optimal therapeutic effects because of enhanced tumour uptake of radioactivity after TMZ, which was confirmed by therapy responses. Therefore, clinical translation of TMZ treatment prior to PRRT might increase tumour responses in NET patients as well.


Background
Systemic internal radiation therapy using radiolabelled peptides, specifically targeting receptors overexpressed on tumour cells, is an attractive cancer treatment. A high absorbed radiation dose is being delivered to tumours with a tolerable toxicity in normal non-targeted tissues. This so-called peptide receptor radionuclide therapy (PRRT) has been shown to be a most effective treatment option for patients with SSTR-expressing neuroendocrine tumours (NETs) [1][2][3].
Most NETs are slowly proliferating tumours, making them relatively resistant to most chemotherapeutics. In more than 50 % of the patients, NETs are diagnosed at a relatively late stage, often with metastatic spread [4], which leaves little chance for curative surgery. Then, PRRT with radiolabelled SST analogues appeared to be an attractive cancer treatment. A high absorbed dose is being delivered to tumours with a tolerable toxicity in normal non-targeted tissues [1][2][3]. Radiolabelled SST analogues, such as 111 In-DTPA-octreotide ( 111 In-octreotide) or 177 Lu-DOTA,Tyr 3 -octreotate ( 177 Lu-TATE), show a high binding affinity for the somatostatin receptor type 2 (SSTR2), which is overexpressed on the vast majority of NETs [5].
PRRT has proven to be an effective therapy in most patients with NETs [1,6]. Nevertheless, despite the improvement of progression-free survival and quality of life, complete remissions after PRRT are still rare. A combination of PRRT with radio-sensitizing chemotherapeutics, such as capecitabine (5-fluorouracil) or temozolomide (TMZ), might induce synergistic effects [7][8][9]. Currently, phase II clinical trials combining PRRT, using 177 Lu-TATE and capecitabine or both capecitabine and TMZ, are ongoing. The first analyses showed promising results [8,10].
When PRRT with chemotherapeutics are being combined, also potential negative interactions between the two therapies should be taken into consideration. Chemotherapy might affect the SSTR2 expression on tumour cells, on which PRRT depends for targeting of radionuclides, for example [11,12]. Moreover, tumour vasculature might be affected by chemotherapy whereas efficient/adequate tumour perfusion is required for local delivery of 177 Lu-TATE [13]. To obtain an optimal therapeutic scheme for the combination of PRRT and TMZ, information on the aforementioned tumour characteristics during treatment is required, which was the aim of this pre-clinical study using an H69-tumour-bearing mouse model. Tumour perfusion was monitored by dynamic contrast-enhanced (DCE)-MRI and SSTR expression by 111 In-octreotide SPECT. The results from these imaging studies were applied in the design of different therapeutic schemes to stratify maximum therapeutic efficacy using TMZ and PRRT.

Experimental setup
We have started a pilot study in which H69 tumourbearing mice were treated with increasing amounts of 177 Lu-TATE to determine which dose resulted in clear tumour responses but without complete cure to allow detection of additional effects during combination therapy. Three groups (n = 7) of H69-tumour-bearing mice were treated with 10, 20 or 30 MBq 177 Lu-TATE and compared with untreated control animals. Monitoring of body weight and tumour size was performed until day 28 after the start of treatment.
For TMZ treatment, different dosing schedules have been reported, including a total administered dose of >1000 mg/kg [14] or the administration of one or more cycles during 5 days, in which a dose of 50 or 120 mg/kg/day was administered [15,16]. Since we planned to administer only one cycle, we used a dose of 50 mg/kg/day for 2 weeks, to study the effects of TMZ treatment in the H69 model, supplemented with investigations into combination of PRRT and TMZ. In this study, the earlier determined optimal dose of 30 MBq 177 Lu-TATE was used.
Imaging studies to determine tumour characteristics during PRRT or TMZ treatment Three groups of mice were included in this imaging study ( Table 1). The PRRT group received a single dose of 50 MBq 177 Lu-TATE on day 1, and the TMZ group was treated with 50 mg/kg TMZ administered orally for 2 weeks, starting at day 0. The control group received saline. The timeline for these studies is shown in Fig. 1a.
PRRT group Mice (n = 4) received a baseline DCE-MRI scan 1 day prior to administration of 50 MBq 177 Lu-TATE. During follow-up of the therapy, MRI scans were acquired on days 4, 7, 11, 15, 20 and 28. Moreover, a SPECT/CT scan to determine the tumour uptake of 177 Lu-TATE was performed 24 h after administration.
TMZ group Mice (n = 7) received one baseline DCE-MRI scan 2 days prior start of TMZ treatment, which was repeated once weekly during and after TMZ treatment until day 28. To determine the level of radiopeptides uptake, 111 In-octreotide SPECT/CT scanning was performed 1 day prior start of TMZ treatment and repeated once weekly for 6 weeks during and after TMZ. At day 1 (before TMZ), day 14 (after TMZ) and day 28, when tumours showed maximal response, three mice where sacrificed to collect the H69 xenograft to determine the level of SSTR expression by in vitro autoradiography using 111 In-octreotide.

Combination therapy of PRRT and TMZ
Seven groups of mice were included to compare H69 responses after different combination treatment schedules, including a placebo control group (receiving oraplus solution without TMZ p.o. once daily.) ( Table 2). Uptake of 177 Lu-TATE was quantified in three to six tumours of each group after PRRT by SPECT imaging. The timeline for these studies is shown in Fig. 1b.

Tumour cell line
The SSTR2-expressing human small cell lung cancer cell line H69 was obtained from ECACC (Salisbury, UK) and grown in RPMI medium (Gibco, Invitrogen Corp.,  Breda, The Netherlands) supplemented with 10 % heatinactivated foetal bovine serum.

Animals and tumour model
All animal experiments have been conducted with prior approval of the animal ethics committee of our institution and performed in accordance with Dutch laws. Male NMRI nu/nu mice (body weight~33 g) were obtained from Harlan (Heerlen, the Netherlands). One week after arrival, at the age of 5-7 weeks, mice were inoculated subcutaneously with 10 7 H69 cells in 0.2-ml HBSS. For all experiments, animals were randomized into matching treatment groups regarding tumour size at the start of treatment 4 weeks after tumour inoculation. Randomized treatment groups were created by matched pairs randomization. Three times a week, mice were weighed and length and width of the tumour were measured using a calliper by a person blinded for the treatment. Tumour volume was calculated using the formula 0.5(length × width) 1.5 adapted from the SWOG criteria. Mice were euthanized when >10 % loss of body weight (BW) since start of the experiment was observed or when tumour volume exceeded 1800 mm 3 .  [17].

SPECT/CT
During scanning experiments, 2.0 % isoflurane/O 2 gas anaesthesia was applied at 0.5 ml/min. Twenty-four hours after injection of 177 Lu-TATE or 111 In-octreotide, helical SPECT/CT of the tumour region was performed with a four-headed NanoSPECT/CT system (BioScan, Washington DC USA) with nine pinhole mice collimators (diameter 1.4 mm) per head. The scans were obtained using 24 projections of 120 s per projection and a quality factor of 0.7. SPECT scans were reconstructed iteratively on a 256 × 256 matrix, using HiSPECT NG software (Scivis, GmbH Göttingen Germany) and ordered subset expectation maximization (OSEM). The total amount of radioactivity (MBq) in the tumour was quantified by 3D quantification using InVivoScope software (IVS, Bioscan, Washington DC USA). To achieve accurate quantification, the camera was calibrated by scanning a 20-mL polypropylene tube phantom filled with a known amount of 177 Lu or 111 In radioactivity. During scanning, body temperature of the mice was maintained using a heated bed.

MRI
Imaging was performed on a pre-clinical 7.0T scanner (Discovery MR901, Agilent technologies/GE Healthcare) with the standard imaging gradient set (300 mT/m, slew rate of 1000 mT/m/ms-rise time = 300 μs). A 150-mm transmit coil and four-channel surface coil (5 cm FOV) were used to acquire all images. DCE-MRI data were acquired following a bolus i.v. injection of Gadobutrol (Gadovist; Bayer Healthcare, Germany). The acquisition parameters used were repetition time (TR) = 10 ms, echo time (TE) = 2 ms, flip angle 12°, matrix size = 116 × 116 and slice thickness = 0.8 mm with a total imaging volume of

Image analysis
Analysis of the SPECT data: the concentration of activity in the tumours was calculated by dividing the total amount of activity in the tumour determined by 3D quantification with HiSPECT NG software through the 3D tumour volume determined during this quantification.
For analysis of the DCE-MRI data, we used two methods: semi-quantitative and quantitative analysis. All calculations were performed using Matlab (Mathworks co.). In semi-quantitative analysis, the signal-intensity time curves were used to determine the time to peak (TTP), maximum signal enhancement (Smax) and the area under the curve (AUC) for the total curve and also for the first 60 s (AUC60). The AUC and AUC60 were calculated using a triangulation algorithm. Furthermore, the wash-in was calculated as a linear slope from the first point before contrast enhancement to the first maximum. The wash-out was calculated from the slope of the first maximum to the measurement at 60 s after injection. The wash-out curve was not necessarily linear in shape, but using this method, we were able to discriminate the wide variety in wash-out kinetics. In order to calculate quantitative parameters from the DCE-MRI data, the signal-intensity time curves were converted to contrast-concentration time curves using a T1-map calibration. A pharmacokinetic compartment model could be fitted to these contrast-concentration time curves. Quantitative analysis consisted of calculation of k trans and k ep , calculated according to Tofts perfusion model [19]. A population-based average arterial input function (AIF) obtained from Weidensteiner [20] was used for consistency in the results. For the response to therapy, comparison of the tumour perfusion parameters was performed; regions of interest were drawn around the tumours, and mean parameter values were calculated. k trans is the calculated rate constant for the passage of contrast agent into the tissue. The rate constant k ep is the reflux of contrast agent from tissue back to the blood. The results presented show the k trans which performed slightly better than the k ep in this model due to the limited amount of wash-out in the 10-min scantime.
In vitro autoradiography on tumours treated with TMZ during the imaging study Sections of 10 μm were sliced from frozen tumours (Cryo-Star HM 560 M; Microm, Walldorf, Germany), mounted on Superfrost plus slides (Menzel, Braunschweig, Germany) and incubated with 10 −9 M 111 Inoctreotide with and without an excess (10 −6 M) of unlabelled octreotide. Adjacent sections were stained with haematoxylin/eosin. Tumour sections were exposed to SR phosphor imaging screens (Packard Instruments Co., Meriden, USA) in X-ray cassettes. After 48-h exposure, screens were read by a Cyclone phosphor imager and analysed using OptiQuant 03.00 (Perkin Elmer, Groningen, The Netherlands). Autoradiogram quantification was expressed in density light units (DLU) per square millimetre.

Statistics
Prism software version 5.0 (Graph Pad) was used to analyse tumour growth and determine statistical significance between groups. One-way ANOVA was used for statistical analysis of tumour uptake, and results are given as mean ± SD. A log rank test was performed for curve comparison.

Lu-TATE dose-response study
All three 177 Lu-TATE groups showed delay of tumour growth compared with the control group (Fig. 2). As expected, the most significant tumour growth suppression was found in the 30 MBq group, although complete response was not reached in any mouse. Imaging studies Effect of single-agent treatment on tumour growth Treatment with either 177 Lu-TATE or TMZ resulted in transient reduction in tumour size/volume, confirming previous results. However, the kinetics of the tumour response after the two treatments differed (Fig. 3a).
The average 177 Lu uptake in the four tumours in the PRRT group was 1.7 ± 0.1 kBq/mm 3 (3.4 ± 0.2 %IA/g) at 24 h after administration of 50 MBq 177 Lu-TATE. Maximum reduction of tumour volume was reached at~day 14, after which the tumours increased in size again. After an average of 43 days, euthanasia was needed because the maximal allowed tumour size (1800 mm 3 ) was reached.
In mice treated with TMZ, the tumour volumes increased until day 9 reaching maximum tumour sizes, after which decrease of tumour sizes was observed. At day 51, at the end of the imaging study, the average tumour volume increased again; however, at that time, none of the tumours exceeded a volume of 1800 mm 3 .

Tumour perfusion during 177 Lu-TATE or TMZ treatment
Tumours subjected to TMZ and 177 Lu-TATE initially showed a fast decrease in k trans , reaching minimum values at day 4 ((2.2 ± 2) × 10 −3 min −1 for 177 Lu-TATE) or day 5 ((2.0 ± 2) × 10 −3 min −1 for TMZ). When tumour size decreased, an increase in k trans was measured. The local maximum in k trans values was reached approximately 2 weeks after (start of ) treatment. Tumours in the group receiving TMZ showed the highest k trans values at day 13 ((11 ± 1) × 10 −3 min −1 ); the 177 Lu-TATE group's maximum k trans value was reached at day 15 ((5.8 ± 2) × 10 −3 min −1 ). During further follow-up, the k trans values decreased again in both the TMZ and 177 Lu-TATE group.
Therapy study Responses to the different treatment schedules As expected, all mice in the control group had to be euthanized due to excessive tumour growth. This was also true for the mice in the single-treatment groups with 177 Lu-TATE (groups 1 and 2) or TMZ (group 3), albeit with a significant delay of 20-50 days (Fig. 5b) In all three groups treated with a combination of 177 Lu-TATE and TMZ, an additive effect of the two treatments was determined. Since in several mice complete tumour response was reached, MST could not be determined. Regarding survival, the difference between the best performing single-therapy group 3 (TMZ) and group 4 (TMZ 14 days after 177 Lu-TATE) was significant (p = 0.046), whereas the other combination groups performed even better. Thus, the applied treatment schedules in these three groups had a striking impact on tumour response. In group 4, still 57 % of tumours reached a volume of 1800 mm 3 before the end of the study at day 123. When 177 Lu-TATE and TMZ were combined from the start (group 5), only 25 % of tumours reached this maximum size. By far, the best result was obtained in group 6 with 177 Lu-TATE administered after 14 days TMZ treatment. Only one tumour (10 %) escaped from treatment, whereas the other tumours showed complete response.

Discussion
Despite the fact that significant tumour response rates have been reported after treatment of NET patients with 177 Lu-TATE or [90Y-DOTA]-D-Phe1-Tyr3-octreotide (90Y-DOTATOC) [21,22], complete responses are still rare, urging for improvements of PPRT protocols. TMZ is currently being applied in combination with radiotherapy for treatment of glioblastoma patients [23]. TMZ (+ cap) is also effective in NET patients [24,25]. Therefore, we examined in mice if this alkylating agent offered synergistic effects when combined with 177 Lu-TATE PRRT.
In this pre-clinical study, we aimed to elucidate the most effective treatment schedule of the TMZ and PRRT combination. We applied a TMZ dose of 50 mg/kg daily for 2 weeks, which appeared to be an effective dose in The human small cell lung cancer (SCLC) H69 cell line, used in our studies, expresses SSTR2 in high densities. As a subtype of NETs, SCLC accounts for about 20 % of all lung cancers [26] and shows a 5-year survival rate of only 20 % [26]. For this type of cancer, there is a clear need to find therapeutic methods resulting in better survival. Several chemotherapeutics, including TMZ, are being used to treat SCLC [27], making the H69 cell line a suitable model for pre-clinical in vivo PRRT research using radiolabelled SST analogues [28]. To study dosimetry and tumour response, Schmitt et al. applied several activity doses of 177 Lu-TATE for PRRT of H69 tumour-bearing mice [28,29]. In their study, tumours Lu-TATE was i.v. administered at day 0 to mice in groups 1, 4 and 5, at day 14 for group 6 and both at day 0 and day 14 for group 2. TMZ was administered orally once daily for 14 days at a dose of 50 mg/kg from day 0 to mice in groups 3, 4 and 6, starting at day 14 for group 5. Eight to ten mice/group. c Average tumour radioactivity: orange bar: groups 1, 2 (first dose), 4, 5; red bar: group 6; green bar: group 2 (second dose). Numbers, respectively, 14, 6 and 6 mice/group showed a significantly higher uptake of radioactivity compared with all normal organs [28]. In a therapy study, using substantially larger tumours than in our study (respectively, 1000 and 2000 mm 3 at day 0), increasing growth delays using doses of 45, 60 and 120 MBq or two fractions of 45 MBq 177 Lu-TATE were found [29].
In the clinical trials in which PRRT was combined with capecitabine and TMZ, the focus mainly has been on the radio-sensitizing effects of the chemotherapeutics [8,30], not considering other effects these agents might have on tumour characteristics. In our opinion, the potential influence of chemotherapeutics on tumour perfusion and SSTR2 expression is of major importance as well, to determine an optimal scheme for the combination of PRRT with chemotherapeutics.
In this study, we measured an increase of H69-tumour perfusion after TMZ treatment. The highest values after 14 days of TMZ treatment coincide with a peak in the uptake of radiolabelled octreotide. The latter observation was confirmed during therapy, as 177 Lu-uptake in H69 tumours after 177 Lu-TATE administration was significantly higher when mice were pre-treated for 14 days with TMZ. This was not the case when pre-treatment consisted of an additional dose of 177 Lu-TATE.
This increased tumour uptake/retention of radioactivity after TMZ treatment can be explained by different phenomena. A relation between tumour volume and concentration of radioactivity has been described, indicating an increased concentration of radioactivity in smaller tumours [28]. In our studies, however, the average tumour size was in the same range at day 0 and at day 13, respectively, a mean tumour volume of 456 ± 122 versus 442 ± 66 mm 3 , demonstrating that the peak of radioactivity uptake at day 13 was not influenced by tumour volume. At later time points, when tumour sizes continued to decrease as a result of TMZ treatment, also the concentration of 111 In-octreotide decreased.
An increased level of SSTR2 expression might also explain the peak of radiopeptide uptake after 14 days of TMZ. Using in vitro studies, Fueger et al. showed increased SSTR2 expression in pancreatic tumour cells 4 days after exposure to several chemotherapeutics, not including TMZ [11,12]. This phenomenon could not explain our findings in the current study, however, as we found no differences in SSTR2 expression prior to or after TMZ treatment, when in vivo H69-uptake of 111 In-octreotide peaked.
Therefore, increased tumour perfusion caused by TMZ treatment offers the best explanation for the raised levels of 111 In and 177 Lu tumour uptake, as was explored and confirmed by application of measurement of perfusion parameters by DCE-MRI.
In multiple studies, TMZ, when administered in a metronomic (repetitive, low dose) schedule, has been demonstrated to have anti-angiogenic effects that may result in normalized tumour vasculature depending on the administered TMZ dose [31,32]. Therefore, direct or indirect anti-angiogenic effect of chemotherapeutics can improve tumour perfusion and reduce the interstitial pressure [33], enabling enhanced delivery to the tumour of therapeutic compounds, including radiolabelled peptides. Our measurements with DCE-MRI support the suggestion that there is an important relationship between functional vasculature and radiopeptide uptake [34].
Considering the anti-tumour response, a small but not significant (p = 0.28) difference was found between group 5 (concurrent 177 Lu-TATE and TMZ) and group 6 ( 177 Lu-TATE after 14 days TMZ treatment) regarding CR, with 75 versus 90 % CR, respectively. However, when we compare group 6 with group 4 (the group in which 177 Lu-TATE was administered at day 1 and TMZ treatment started at day 14), a striking and significant difference (p = 0.046) between those groups was found, in agreement with the fact that TMZ potentiates the effect of 177 Lu-TATE.
Therefore, the optimal response found in the group receiving 177 Lu-TATE after TMZ treatment might be based on a combination of factors: 1. Increased uptake of 177 Lu-TATE because of enhanced perfusion resulting in a higher absorbed tumour radiation dose and therefore an increased anti-tumour response [35]. 2. Increased radiosensitivity induced by TMZ. The cytotoxicity of TMZ is primarily due to alkylation at the O6 position of guanine combined with an additional alkylation at the N7 position. TMZ has been reported to act synergistically with radiation therapy, as alkylated guanine has been reported to be radio-sensitizing [9], as was shown in multiple types of cancer cells [9,36]. As the half-life of 177 Lu is 6.7 days [28], also the group receiving 177 Lu-TATE at the same day that TMZ treatment was started might have had some benefit from the radiosensitizing effects of TMZ. 3. Increased oxygenation might have contributed to improved therapeutic responses, as was demonstrated for radiation therapy [37]. After betaparticle radiation, DNA damage occurs as a result of the creation of free oxygen radicals [38], so the effect of 177 Lu-TATE treatment might be enhanced when tumour oxygenation is improved. The mice receiving 177 Lu-TATE 14 days prior to TMZ treatment did not have this benefit.
Despite the fact that our model has been carefully chosen, it has important differences from the patient situation. Differences in, e.g. tumour growth rate might influence the outcome of these studies in a clinical setting. Therefore, the translational value of our findings into the clinic remains to be proven.
During our combination therapy study, no signs of severe toxicity like, e.g. remarkable loss of weight have occurred. Nevertheless, during PRRT as well as chemotherapy, toxicity is an aspect of great concern. In a clinical study, a treatment scheme in which PRRT was combined with capacetabine and TMZ has been shown to be safe [8]. Considering increased uptake of radionuclides after TMZ treatment, in our opinion, this is likely to be solely applicable to the tumour because of its immature vasculature. However, further investigation on toxicity and uptake of radionuclides by the normal organs is warranted.

Conclusions
The use of SPECT/CT and MRI enabled us to optimize a treatment schedule for the combination of PRRT and TMZ in H69-xenografted nude mice. Increased perfusion following TMZ pre-treatment, determined by DCE-MRI, coinciding with increased uptake of 177 Lu-TATE as proven by SPECT/CT, was reflected by the induced therapeutic effects we observed.