Dosimetric comparison of radionuclides for therapy of somatostatin receptor-expressing tumors

Int. J. Radiation Oncology Biol. Phys., Vol. 51, No. 2, pp. 514 –524, 2001 Copyright © 2001 Elsevier Science Inc. Printed in the USA. All rights reserved 0360-3016/01/$–see front matter

PII S0360-3016(01)01663-7

BIOLOGY CONTRIBUTION

DOSIMETRIC COMPARISON OF RADIONUCLIDES FOR THERAPY OF SOMATOSTATIN RECEPTOR–EXPRESSING TUMORS PETER BERNHARDT, PH.D.,* SVEN ANDERS BENJEGÅRD, M.SC.,* LARS KO¨ LBY, M.D., PH.D.,† VIKTOR JOHANSON, M.D., PH.D.,‡ OLA NILSSON, M.D., PH.D.,† HÅKAN AHLMAN, M.D., PH.D.,‡ AND EVA FORSSELL-ARONSSON, PH.D.* Departments of *Radiation Physics, †Surgery, and ‡Pathology, Lundberg Laboratory for Cancer Research, Sahlgrenska University Hospital, University of Go¨teborg, Go¨teborg, Sweden Purpose: Therapy of tumors expressing somatostatin receptors, sstr, has recently been clinically tested using somatostatin analogues labeled with 111In and 90Y. Several other radionuclides, i.e., 131I, 161Tb, 64Cu, 188Re, 177 Lu, and 67Ga, have also been proposed for this type of therapy. The aim of this work was to investigate the usefulness of the above-mentioned radionuclides bound to somatostatin analogues for tumor therapy. Methods: Biokinetic data of 111In-labeled octreotide in mice and man were used, primarily from our studies but sometimes from the literature. Dosimetric calculations were performed with the assumption that biokinetics were similar for all radionuclides bound to somatostatin analogues. The cumulated tumor:normal-tissue activity ˜ , was calculated for the various physical half-lives of the radionuclides. Using mathematical concentration, TNC ˙ , and tumor:normal-tissue mean absorbed models, the tumor:normal-tissue mean absorbed dose rate ratio, TN D dose ratio, TND, were calculated for various tumor sizes in mice and humans. ˜ of radionuclide-labeled octreotide increased with physical half-life for most organs, both in mice Results: TNC ˙ showed that radionuclides emitting electrons with too high energy are not suitable for and in humans. TN D therapy of small tumors. Furthermore, radionuclides with a higher frequency of photon emissions relative to ˙ and are thus less suitable for therapy than radionuclides with a lower electron emissions will yield lower TN D frequency of photon emissions. The TND was highest for 161Tb in both mice and humans. Conclusions: The results demonstrate that long-lived radionuclides, which emit electrons with rather low energy and which have low frequency of photon emissions, should be the preferred therapy for disseminated small sstr-expressing tumors. © 2001 Elsevier Science Inc. Dosimetry, Therapy, Somatostatin analogues, Neuroendocrine tumors, Radionuclides.

expressing tumors (e.g. 11). The high 111In activity concentration in such tumors after i.v. injection of 111InDTPA-D-Phe1-octreotide has led to attempts with systemic radiotherapy using this compound (2, 11–13). Several other radionuclides bound to somatostatin analogues have been tried for therapy of sstr-expressing tumors. Studies in vitro and in rats have shown that 161Tb-DTPA-D-Phe1octreotide is a promising radiopharmaceutical both for intraoperative scintillation detection and radiation therapy (4). Recent animal studies using 90Y-DOTA-octreotide have shown complete regression of rat pancreatic tumors (14). Also, in clinical trials 90Y-DOTA-octreotide or 90Y-DOTA-lanreotide reduced the growth rate of tumors and caused tumor regression in individual cases (15, 16). The ␤-emitter 131I has recently been bound to the somatostatin analogue Woc-3a and administered i.v. in therapeutic amounts (59 GBq) to a patient (1). Using intratumoral injection of the radiolabeled somatostatin

INTRODUCTION Neuroendocrine tumors frequently overexpress somatostatin receptors (sstr). Somatostatin analogues, e.g. octreotide, lanreotide, and Woc3a, have been labeled with the following radionuclides: 131I (1), 111In (2), 90Y (3), 161Tb (4), 64Cu (5), 188Re (6), and 67Ga (7). After i.v. injection of these radiolabeled somatostatin analogues in animals or humans, high activity concentrations in sstr-expressing tumor tissue compared to normal tissue were found, suggesting a potential therapeutic role of these radiopharmaceuticals. Initially 123I-Tyr3-octreotide was used to visualize endocrine pancreatic tumors by single photon emission computed tomography (SPECT) (8). The results indicated a potential use of 131I-labeled octreotide for systemic radiation therapy of sstr-expressing tumors (9, 10). Today, 111InDTPA-D-Phe1-octreotide has become the dominant radiolabeled somatostatin analogue for visualization of sstrReprint requests to: Peter Bernhardt, Department of Radiation Physics, Go¨teborg University, Sahlgrenska University Hospital, S-413 45 Go¨teborg, Sweden. Tel: ⫹46-31-3421916; Fax: ⫹4631-822493; E-mail: [email protected] This work was supported by grants from the Swedish Cancer

Society (2998, 3427) and the Swedish Medical Research Council (5220, 6534). Received Nov 6, 2000, and in revised form May 8, 2001. Accepted for publication May 10, 2001. 514

Radionuclides for tumor therapy

analogue 188Re-RC-160, Zamora et al. (6) obtained therapeutic effects on sstr-expressing tumors in nude mice. High uptake of 64 Cu-TETA-Y3-TATE in sstr-expressing tumor tissue was recently shown in vitro and in vivo (5). Furthermore, biodistribution studies in rats have demonstrated high uptake of 67 Ga-DFO-octreotide in sstr-expressing tissues (7), and 177Lu has also been suggested as a suitable radionuclide for this type of therapy (17). One of the limiting factors in radionuclide therapy is the absorbed dose to normal tissues. The radionuclides mentioned above emit either electrons or positrons together with photons. The electrons and positrons will deposit their energy locally (within a few nm to mm from the site of decay). On the other hand, the photons will contribute to the whole-body irradiation, which should be kept low in radiation therapy. Most of the biokinetic data on radiolabeled somatostatin analogues in man have been collected up to 72 h after injection (18 –20). To predict the therapeutic efficacy of more long-lived radionuclides, long-term biokinetics are needed. We have therefore followed the biokinetics up to 22 days both for tumor and normal tissues in a patient with metastatic midgut carcinoid, diagnosed and treated with 111 In-DTPA-D-Phe1-octreotide (11, 21). We have also studied the biokinetics of 111In-DTPA-D-Phe1-octreotide 0.5 h to 14 days after injection into mice bearing a GOT1 tumor xenograft derived from this patient (22, 23). The aim of this study was to compare the dosimetry of the proposed radionuclides bound to somatostatin analogues for radiotherapeutic purposes. Human biokinetic data of 111Inlabeled octreotide from our own patients and the literature (cf. 18 –21, 24) were used, together with long-term biokinetics of 111In-DTPA-D-Phe1-octreotide in GOT1-bearing nude mice (23). METHODS AND MATERIALS Computer simulations From the biokinetic data, the tumor:normal-tissue activity concentration ratio, TNC, was obtained as

515

was calculated for different radionuclides in Eq. 1:

TNC˜ ⫽

冕 冕

C T共t兲 䡠 e ⫺␭tdt

C N共5兲 䡠 e

˙ N共t兲 ⫽ C N共t兲共 D

冘E n ␾ i

e,i e,i

⫹ C T共t兲

mT mN

i

e,i e,i

T,e,i

i

e,i e,i

i

p,i p,i

i

p,i p,i

p,i p,i

N,p,i

兲

N,p,i

⫺ ␾ T,p,i兲

(2)

冘E n ␾ ⫹ 冘E n ␾ 兲 ⫹ C 共t兲共 冘 E n ⫹ 冘 E n ␾ 兲 (3) e,i e,i

i

i

T,e,i

e,i e,i

i

i

p,i p,i

p,i p,i

T,p,i

N,p,i

where ␾T,e,i isthe mean absorbed electron energy fraction in the tumor. For tumors ⱕ1 g, the absorbed fraction in the tumor from photons in the tumor was neglected. The tumor:normal-tissue mean absorbed dose rate ratio, ˙ , is as follows in Eq. 4: TN D

p,i p,i

i

e,i e,i

T

i

i

where Ee,i and Ep,i are the energies of the emitted electrons and photons, respectively, per transition i, ne,i and np,i are the numbers of emitted electrons and photons, respectively, per transition, ␾N,e,i and ␾N,p,i are the mean absorbed electron and photon energy fractions, respectively, per transition, mN and mT are the mass of the normal and tumor tissue, respectively, and ␾T,p,i is the mean absorbed photon energy fraction in the tumor. In general, the deposition of the electron energy was assumed to take place at the decay site; i.e., the fraction of the electron energy absorbed in the normal tissue, ␾N,e,i, was 1. The mean absorbed fraction in the normal tissue from electrons originating from the tumor was neglected. ˙ T(t), is then The mean absorbed dose rate to the tumor, D (Eq. 3):

冘E n ␾ ⫹ 冘E n ␾T,p,i兲 ⫹ 冘E n ⫹ 冘E 冘E n ⫹ 冘E n ␾ ⫹ TNC mm 冘E n 共␾ ⫺ ␾

共TNC ⫺ 1兲共

(1)

dt

冘E n ␾ 冘E n 共 ␾

⫹

N,e,i

N

for various normal tissues, where CT(t) and CN(t) are the decay-corrected activity concentrations (% of administered activity per gram [%IA/g]) in the tumor, T, and normal, N, tissue, respectively, at time post administration. The cumulated tumor:normal-tissue activity concentration ratio, TNC˜,

⫺␭t

C˜ T ⫽ ˜ CN

where ␭ is the physical decay constant for the radionuclide studied. The tumor:normal-tissue mean absorbed dose rate ratio, ˙ , was estimated both for mouse and man. The body TN D was simulated by an ellipsoid representing normal tissue. Tumors were modeled as spheres and randomly distributed throughout the ellipsoid. The distribution of radionuclides in the spheres and in the ellipsoid was assumed to be uniform. The mean absorbed dose rate to the normal tissue, ˙ N(t), for uniformly distributed activity in the ellipsoid was D as follows (Eq. 2):

˙ T共t兲 ⫽ 关C T共t兲 ⫺ C N共t兲兴共 D

C T共t兲 TNC ⫽ C N共t兲

˙ ⫽ TND

● P. BERNHARDT et al.

N,p,i

N

i

p,i p,i

N,p,i

i

n ␾ N,p,i

p,i p,i

(4) T,p,i 兲

Fig. 1. The calculated cumulated tumor:normal-tissue activity concentration ratio, TNC˜, vs. the physical half-life of the radionuclide for the following: (a– c) i.v.-injected 111In-DTPA-D-Phe1octreotide in nude mice bearing GOT1 tumor (23) and (d) i.v.-injected 111In[DOTA,Tyr3]octreotide in patients with neuroendocrine tumors (19). The half-lives of the studied radionuclides are indicated.

Radionuclides for tumor therapy

The tumor:normal-tissue mean absorbed dose ratio, TND, was calculated for the whole body (Eq. 5):

● P. BERNHARDT et al.

517

The mean absorbed photon and electron energy fractions The mean absorbed photon energy fractions, ␾N,p,i, for

冘E n ␾ ⫹ 冘E n ␾ 兲 ⫹ 冘E n ⫹ 冘E n ␾ 冘E n ⫹ 冘E n ␾ ⫹ TNC˜mm 冘E n 共␾ ⫺ ␾ 兲

共TNC˜ ⫺ 1兲共 TND whole bod y ⫽

e,i e,i

i

T,e,i

i

p,i p,i

T,p,i

i

e,i e,i

i

p,i p,i

N,p,i

(5)

T

i

e,i e,i

p,i p,i

i

N,p,i

N

Equation 5 is then valid for a homogeneous radionuclide distribution in the whole body. We calculated the TND for an organ, assuming that the photons irradiating the organ and tumor originate from a homogeneous radionuclide distribution in the whole body (Cw[t]), instead of a single tissue. TNDorgan is as follows (Eq. 6):

共C˜ T ⫺ C˜ N兲共 TND organ ⫽

i

p,i p,i

N,p,i

T,p,i

a 70-kg (human) or 30-g (mouse) ellipsoid, with principal axes forming the ratio 1:1.8:9.27, were taken from MIRD pamphlet No. 3 (25) and No. 8 (26), respectively. The decay properties of the radionuclides were taken from the World Wide Web table of radioactive isotopes (27). The mean absorbed electron energy fraction totumors, ␾T,e,i,

冘E n ␾ ⫹ 冘E n ␾ ⫹ 冘E n ␾ 兲 ⫹ C˜ 冘E n ⫹ C˜ 冘E n ␾ m C˜ 冘 E n ⫹ C˜ 冘 E n ␾ ⫹ C˜ 冘E n 共 ␾ ⫺ ␾ 兲 m i

e,i e,i

T,e,i

i

p,i p,i

T,e, j

i

p,i p,i

T,p,i

N

i

e, j e, j

W

i

p,i p,i

N,p,i

(6)

T

N

e,i e,i

i

W

i

p,i p,i

N,p,i

T

N

i

p,i p,i

N,p,i

T,p,i

Table 1. Some physical properties of radionuclides proposed for therapy using radiolabeled somatostatin analogues Radionuclide 64

Half-life 12.7 h

Cu

Decay mode

Electron energy (keV)

Abundance (%)

p/e

␤, EC, ␤⫹

190 278 0–1 7–10 84–92 935 0–4 19–22 145–167 219–242 191 46–97 250–360 1.5–8.0 17–26 40–49 135 154 180 47–48 102–113 149 728 795

37 18 169 61.3 31 100 293 16 10 6 89 14 2 254 51 54 23 67 10 17 18 79 25 72

1.5

67

3.26 days

EC

90

2.67 days 2.83 days

␤ EC

131

8.04 days

␤

161

6.91 days

␤

177

6.7 days

␤

Ga Y In

111

I Tb

Lu

188

Re

17 h

p/e ⫽ photon: electron energy ratio.

␤

4.5 1.8 10⫺6 11

2 0.18

0.24 0.07

518

Radionuclides for tumor therapy

was calculated using the method of Howell et al. (28), which uses the experimental determined energy range relationship by Cole (29). The calculation of the mean absorbed fraction by the tumor, ␾T,e,i, was carried out for different tumor sizes and electron energies. The tumor sizes ranged from one single cell with a mass of 1 ng (Rc ⫽ 6.2 ␮m) to a tumor of 1 kg (Rc ⫽ 6.2 cm). All calculations were made with a program written in IDL (Research Systems Inc., USA) and were carried out on a Macintosh Quadra 950 computer. Biokinetic data of 111In-DTPA-D-Phe1-octreotide in nude mice, transplanted with a well-differentiated carci˜ noid, GOT1 (cf. 22), were used in the calculation of TNC and TND for the mouse (23). The 111In activity concentration in various tissues (tumor, blood, thigh muscle, adrenals, omental fat, heart, small intestine, kidneys, liver, lungs, salivary glands, spleen, pancreas, and whole body) was obtained at 0.5 h to 14 days after i.v. injection of the radiopharmaceutical. For each tissue, bi- or threeexponential time activity curves, C(t), were fitted to the 111 In activity concentration for each tissue, corrected for physical decay. ˜ in man, the mean biokinetic data of To estimate TNC 111 In-DOTA-D-Tyr3-octreotide from 18 patients were used (19). For each tissue, bi-exponential time activity curves, C(t), were fitted. The TNDwhole body was calculated for both man and mice. For man, the mean whole-body activity concentration from 13 patients injected with 111In-DTPA-D-Phe1-octreotide (18) was used, together with biokinetic data on the tumor tissue from the patient with midgut carcinoid tumor from which the GOT1 cell line was derived (21). The TNDbone marrow in man and mice was calculated with the assumption that the activity concentration in bone marrow was similar to that in blood. The blood activity concentrations from Krenning et al. (24) were used for man. The small dimensions of the bone cavities in mice mean that only a small energy fraction of highenergy electrons will be absorbed in the red marrow. Therefore, for bone marrow in mice, we have also used the absorbed fractions 10% and 46% for the electrons from 90Y and 131I, respectively (30). ˙ were calculated for the radionuclides TND and TN D 131 I, 111In, 90Y, 161Tb, 177Lu, 64Cu, 188Re, and 67Ga. The physical properties of the radionuclides used in the calculations are listed in Table 1.

● P. BERNHARDT et al.

519

RESULTS TNC˜ was calculated based on the biokinetics of 111InDTPA-D-Phe1-octreotide in nude mice carrying the human ˜ increased with carcinoid tumor GOT1 (Fig. 1a– c). TNC increasing physical half-life of the radionuclide. Similar results were obtained from the published human data on 111 In-DOTA-D-Phe1-octreotide (Fig. 1a– c) (cf. 19). ˙ whole body, TNDwhole body, and The relation between TN D tumor size are given in Fig. 2 for the radionuclides studied. ˙ whole body is rather low for radionuclides emitting a TN D high fraction of energy as photons, e.g. for 111In and 67Ga. The high-energy electron emitters 90Y and 188Re give high ˙ whole body for tumors larger than 1 g. However, for TN D ˙ whole body is reduced because of the low smaller tumors TN D absorbed energy fractions. TNDwhole body was calculated for mice and man. The long-lived radionuclides 161Tb and 131I gave relatively high TNDwhole body, whereas the short-lived 64Cu and 188Re gave low TNDwhole body. Both for man and mouse, TNDwhole body was lower for radionuclides emitting a high abundance of photons and higher for pure electron emitters. Figure 3 shows the TNDbone marrow for mice and man. In Fig. 3c is shown the calculation of TNDbone marrow in mice, given the assumption that only a fraction of emitted electrons will be absorbed in bone marrow. TNDbone marrow for 90Y and 131I are increased if the fractions of electrons absorbed by the bone marrow are decreased. DISCUSSION For successful therapy with radiopharmaceuticals, the absorbed dose in the tumor must be higher than in the normal tissue. The tumor:normal-tissue activity concentration ratio, TNC, varies with time because of different pharmacokinetics in the tissues and physical decay of the radionuclide. This is taken into account in the cumulated ˜. tumor:normal-tissue activity concentration ratio, TNC TNC˜ can accordingly be used to predict the therapeutic potential of a radiopharmaceutical, if the radiation is assumed to be absorbed at the site of decay. Quite often the radionuclides emit not only particles but also photons, which are not locally absorbed. The absorbed fraction of the emitted electrons depends on the electron energy and size of the target. The tumor:normal-tissue absorbed dose ˙ , takes into account the electron energy, rate ratio, TN D the abundance of emitted photons, and the tumor size. The tumor:normal-tissue absorbed dose ratio, TND, takes

˙ wholebody, vs. tumor mass for various Fig. 2. (a) The tumor:normal-tissue mean absorbed dose rate ratio for the whole body, TN D radionuclides in the simulated human model. The TNC is set to 25. (b) The tumor:whole-body mean absorbed dose ratio, TNDwhole body, vs. tumor mass for various radionuclides in a simulated human model. The biokinetic data for tumor are from Andersson et al. (21) and for whole body are from Forssell-Aronsson et al. (18) for all studied radionuclides. (c) The tumor:wholebody mean absorbed dose ratio, TNDwhole body, vs. tumor mass for various radionuclides in a simulated mouse model. The biokinetic data for tumor are from Bernhardt et al. (23) for all studied radionuclides.

520

I. J. Radiation Oncology

●

Biology

●

Physics

Volume 51, Number 2, 2001

Radionuclides for tumor therapy

into account furthermore the pharmacokinetics and physical half-life of the radionuclide. It is important to compare TND values when radionuclides suitable for therapy are to be chosen. However, the radiosensitivity and any dose rate effects are not included in the present models. To analyze the physical properties of radionuclides for therapy of sstr-expressing tumors, the observed long-term biodistribution of 111In-DTPA-D-Phe1-octreotide in the transplantable well-differentiated human carcinoid tumor GOT1 in nude mice was used (23). There are several human pharmacokinetic studies of 111In-DTPA-D-Phe1-octreotide, but to our knowledge only one includes biokinetics for tumor tissue (cf. 21). We also used data observed for 111 In-DOTA-Tyr3-octreotide (19) and 111In-DTPA-Phe1octreotide in man (18). TNC increased immediately after the injection of 111 In-DTPA-D-Phe1-octreotide for all studied organs in the GOT1-bearing nude mice and remained elevated during the time of study (0.5 h to 14 days), except in the spleen, liver, and muscle, where TNC was maximal at 3 to 7 days after injection (23). A similar pattern was seen for TNC of the original GOT1 tumor in the patient (21). This increase in TNC with time indicates a favorable therapeutic situation for long-lived radionuclides. To il˜ vs. physical half-life was lustrate these conditions, TNC calculated (Fig. 1). It was then assumed that the activity concentration in the tumor was not affected by therapeutic effects during the course of the study. This seems to be a reasonable assumption, because release of radioactivity during tumor regression may relate to a similar decrease in tumor mass. The present results clearly show that it is favorable not to use radionuclides with shorter ˜ values half-lives, because this will result in lower TNC 161 131 (Fig. 1). With long-lived Tb and I, 2 to 4 times ˜ values were obtained compared to the more higher TNC ˜ was also seen for short-lived 64Cu and 188Re. High TNC 111 3 In-DOTA-Tyr -octreotide in man (Fig. 1d) (19). However, in the latter study, the biodistribution was only followed for 48 h, a fact that limits its predictive value for a long-lived radionuclide. The absorbed dose in a tumor depends on the radiation range and the tumor size. Electrons and positrons have ranges from tenths of nm to several mm, depending on their energy. For uniform activity distribution in larger

● P. BERNHARDT et al.

521

tumors (⬎1 g), almost all energy of the emitted electrons will be deposited in the target. However, for small tumors (⬍1 g) a small energy fraction from the emitted electrons and positrons will be deposited within the tumor. By calculating the TND rate, i.e., having TNC fixed, it is possible to compare the influence of photon abundance and electron energy for the different radionuclides (Fig. 2a). 90Y and 188Re emit electrons with high energy, ˙ for tumor masses larger than 1 g which gives high TN D and will rapidly decrease with decreasing tumor mass. Radionuclides emitting lower electron energies are thus more suitable for therapy of small tumors. Most radionuclides suggested for sstr-based radiotherapy also emit photons to varying degrees. The photons will deposit their energy in a nonlocalized way and will thus give ˙ values. The low electron energy emitters reduced TN D 111 67 In and Ga also emit numerous photons, as reflected ˙ levels. by their low TN D The biokinetics of radiolabeled octreotide was considered in the mathematical TND model. The short-lived radionuclides had markedly lower TND compared with the long-lived radionuclides (Fig. 2b and 2c). The TNDwhole body was lower in mice than in man, which is partly explained by the high activity concentration in the kidneys of mice (more than half of the activity content). Even if our results demonstrate that long-lived radionuclides are preferred for therapy of sstr-positive tumors, it is known that somatostatin analogues will have different biodistributions and affinities with different chelates and radionuclides (31). There might also be an interpatient variability of the biokinetics that has to be studied further. Therefore, the benefit of using long-lived radionuclides has to be further investigated by comparative studies of different radiolabeled analogues. Furthermore, the use of radionuclides that are too long lived will give low-dose rates that might reduce the biologic efficiency (32). TNDbone marrow was calculated with the assumption that the activity concentration in bone marrow was similar to that in blood. To our knowledge, no detailed study on the pharmacokinetics in bone marrow have been performed for radiolabeled somatostatin analogues. We have measured the 111 In activity concentration in human bone marrow in three surgical samples from patients injected with 111In-DTPA-

Fig. 3. (a) The tumor:bone-marrow mean absorbed dose ratio, TNDbone marrow, human, vs. tumor mass for various radionuclides in a simulated human model. The biokinetic data for tumor are from Andersson et al. (21), for whole body are from Forssell-Aronsson et al. (18), and for blood are from Krenning et al. (24) for all studied radionuclides. (b) The tumor:bone-marrow mean absorbed dose ratio, TNDbone marrow, mouse, vs. tumor mass for various radionuclides in a simulated mouse model. The biokinetic data for tumor are from Bernhardt et al. (23) for all studied radionuclides. (c) The tumor:bone-marrow mean absorbed dose ratio, TNDbone marrow, mouse, vs. tumor mass for various radionuclides in a simulated mouse model. The biokinetic data for the tumor are from Bernhardt et al. (23) for the studied 90Y and 131I. The emitted electrons in bone marrow are assumed to be totally absorbed 90Y and 131I, or only a fraction of the emitted electron energies were assumed to be absorbed. For 90Y, this fraction was 10% (90Y[10%]), and for 131I this fraction was 46% (131I[46%]) according to Muthuswamy et al. (30). In all figures, the bone marrow activity concentration was assumed to be the same as in the blood.

522

I. J. Radiation Oncology

●

Biology

●

Physics

D-Phe1-octreotide and obtained bone marrow:blood ratios of 1.6 –2.7 (33). Therefore, our TNDbone marrow values might be somewhat overestimated for 111In. TNDbone marrow must also be investigated for each radionuclide and labeling method. Given our assumption about similar pharmacokinetics, it was shown that the radionuclides emitting a high abundance of photons gave lower TNDbone marrow in the human model than in the mouse model because of a higher absorbed fraction of photons in the normal tissue. Furthermore, TNDbone marrow was higher for pure electron emitters in man than in mice because of higher TNC (Fig. 3a and 3b). However, the emitted high-energy electrons will not be totally absorbed into the small red bone cavities in mice, in contrast to the bone cavities of man. In mice, Muthuswamy et al. (30) have calculated that the mean absorbed fraction of the emitted electrons in bone marrow from 90Y and 131I was 10% and 46%, respectively. This fact will increase the calculated TNDbone marrow considerably, especially for highenergy electron emitters such as 90Y (Fig. 3c). Therefore, the bone marrow in mice will probably not be such a critical organ for high-energy electron emitters, whereas in man it is (15). The results indicate that radionuclides for therapy of small disseminated tumors should emit few photons, have a rather long half-life (T1/2 ⬎ 2 days), and emit moderate electron energy. Of the radionuclides proposed for sstrbased radiotherapy, it seems that 111In and 67Ga emit too many photons to be optimal. However, both these radionuclides emit Auger electrons, which may enhance the radiobiologic effect if the radionuclides are deposited close to the nucleus of tumor cells (cf. 34). In our study on tumor cell cultures incubated with 111In-DTPA-DPhe1-octreotide and studied by electron microscopic autoradiography, 111In was mainly located in the cytoplasm and also, to some extent, in the nucleus (35). If Auger electrons are to be of therapeutic use, an Auger electron emitter with low amounts of photons and rather low energies of the conversion electrons would be preferred (23). The very high TNC values obtained in patients with advanced neuroendocrine tumors may explain the positive effects (reduced hormonal symptoms and stabilization of tumor growth) observed after 111In-DTPA-DPhe1-octreotide therapies (2, 11–13). No complete tumor remission has been reported in man. The toxicity related to treatment included various grades of anemia and myelosuppression. The toxicity on red blood marrow cells might be reduced by using radionuclides emitting low abundance of photons. The positron emitter 64Cu will give relatively low TND values because of its short half-life. Studies on 64Cu, bound to the somatostatin analogues TETA-OC and TETA-Y3TATE, have been done (5). If 64Cu-TETA-Y3-TATE were used for therapy, it would be limited by low TNC values immediately after injection. Another short-lived radionuclide proposed for radionuclide therapy is 188Re. Intratu-

Volume 51, Number 2, 2001

moral injection of 188Re-RC-160 into xenografted human prostate tumors (6) obtained demonstrable therapeutic responses. For systemic therapy, the short half-life will, however, limit the possibility of reaching high TND values. Furthermore, it will be difficult to treat small, disseminated tumors with this agent, because of the high energy of its electrons. Use of 90Y-DOTA-D-Phe1-[Tyr3]-octreotide (14) achieved complete remission of transplants of the rat pancreatic tumor CA 20948 in rats, without side effects. Also in clinical trials, marked regression of neuroendocrine tumors have been observed (15). Side effects, i.e., renal toxicity, anemia, and thrombocytopenia, were observed. Therapy of sstr-expressing tumors by high-energy electron emitters bound to somatostatin analogues such as 90Y-DOTA-D-Phe1-[Tyr3]-octreotide must be further studied. It will, however, be difficult to treat small, disseminated tumors with this agent because of the high energy of its electrons. The ␤-emitter 131I has recently been bound to the somatostatin analogue Woc-3a and administered i.v. in therapeutic amounts (59 GBq) to a patient (1). The therapeutic amount administered was based on pharmacokinetics obtained from diagnostic scintigraphy (59 MBq 131 I-Woc-3a up to 48 h p.i.). However, the activity stayed longer in the body than the authors had calculated, and the patient died from a myocardial infarction 5 days after administration of the radiopharmaceutical. Biopsy demonstrated coagulation necrosis of both brain and lung tumors. No evidence of radiation damage to the normal tissues could be confirmed. Further trials with 131I-Woc3a have been stopped. De Jong et al. (4) performed distribution studies in rats after administration of 161Tb-DTPA-D-Phe1-octreotide with promising results, but no further studies have since been published with this radionuclide. Very recently, marked tumor reduction has been obtained in animal studies with [177Lu-DOTA, Tyr3]octreotate (36). In our calculations, 161 Tb and 177Lu showed the highest TND values. Therefore, we propose that these radionuclides should be further investigated for therapy using radiolabeled somatostatin analogues in the future. The human carcinoid tumor GOT1 xenografted to nude mice is a novel model for the study of radionuclide therapy ˜ and TND valof sstr-positive tumors. Comparison of TNC ues from our animal and human studies revealed similar results, indicating that the optimal radionuclides should emit few photons and have relatively long half-lives. It was also evident that radionuclides emitting high-energy electrons were not suitable for treatment of small tumors, because of the low absorbed energy fraction in small tumors. However, the differences between mice and humans in regard to TND values of kidneys and bone marrow must be considered when planning and evaluating radionuclide therapy studies in mice.

Radionuclides for tumor therapy

● P. BERNHARDT et al.

523

REFERENCES 1. Espenan GD, Nelson JA, Fisher DR, et al. Experience with high dose radiopeptide therapy: The health physics perspective. Health Phys 1999;76(3):225–235. 2. Krenning EP, Bakker WH, Kooij PPM, et al. Radiotherapy with a radiolabeled somatostatin analogue [111In-DTPAPhe1] octreotide. In: Wiedenmann B, Kvols LK, Arnold R, Rieken EO, editors. Molecular and cell biological aspects of gastroenteropancreatic neuroendocrine tumor disease. Ann N Y Acad Sci 1994;733:496 –506. 3. Otte A, Jermann E, Behe M, et al. DOTATOC: A powerful new tool for receptor-mediated radionuclide therapy. Eur J Nucl Med 1997;24:792–795. 4. de Jong M, Breeman WAP, Bernad BF, et al. Evaluation in vitro and in rats of Tb-DTPA-octreotide, a somatostatin analogue with potential for intraoperative scanning and radiotherapy. Eur J Nucl Med 1995;22:608 – 616. 5. Lewis JS, Srinivasa A, Schmidt MA, Anderson CJ. In vitro and in vivo evaluation of Cu-TETA-Tyr-Octreotate. A new somatostatin analogue with improved target tissue uptake. Nucl Med Biol 1999;26:267–273. 6. Zamora PO, Gulhke S, Bender H, et al. Experimental radiotherapy of receptor-positive human prostate adenocarcinoma with Re-RC-160, a directly-radiolabeled somatostatin analogue. Int J Cancer 1996;65:214 –220. 7. La´znı´ckova´ A, La´znı´cek M, Tretjnar F, Ma¨cke HR. Studies in rats on octreotide labelled with Ga-67 –a potential radiopharmaceutical for the treatment of receptor-positive tumors (IAEA-SR-209/20). In: International Seminar on Therapeutic Applications of Radiopharmaceuticals. Proceedings of a symposium (Hyderbad, 1999). International Atomic Energy Comission, Vienna, 1999. 8. Lamberts S, Bakker W, Reubi J-C, et al. Somatostatin imaging in the localization of endocrine tumours. N Engl J Med 1990;323:1246 –1253. 9. Breeman WAP, Hofland LJ, Bakker WH, et al. Radioiodinated somatostatin analogue RC-160: Preparation, biological activity, in vivo application in rats and comparison with 123I-Tyr3-octreotide. Eur J Nucl Med 1993;20:1089 – 1094. 10. Buxton-Thomas M, Ramage JK, Williams R. Comparison of indium-111-labelled octreotide and I-123-metaiodobenzylguanidine imaging in patient with secondary carcinoid tumours of the liver. Eur J Nucl Med 1994;21(Suppl): 200. 11. Fja¨lling M, Andersson P, Forssell-Aronsson E, et al. Systemic radionuclide therapy using Indium-111-DTPA-DPhe1-octreotide in midgut carcinoid syndrome. J Nucl Med 1996;37(9):1519 –1521. 12. McCarthy KE, Woltering EA, Espenan GD, Cronin M, Maloney TJ, Anthony LB. In situ radiotherapy with Inpentetreotide: Initial observations and future directions. Cancer J Sci Am 1998;4:94 –102. 13. Krenning EP, de Jong M, Kooij PP, et al. Radiolabelled somatostatin analogue(s) for peptide receptor scintigraphy and radionuclide therapy. Ann Oncol 1999;10(Suppl. 2): S23–29. 14. Stoltz B, Weckbecker G, Smith-Jones PM, Albert R, Raulf F, Bruns C. The somatostatin-targeted radiotherapeutic [90Y-DOTA-D-Phe1, Tyr3]octreotide (Y-SMT 487) eradicates experimental rat pancreatic CA 20948 tumours. Eur J Nucl Med 1998;25:668 – 674. 15. Otte A, Hermann R, Heppeler A, et al. Yttrium-DOTATOC: First clinical results. Eur J Nucl Med 1999;26:1439 – 1447. 16. Leimer M, Kurtaran A, Smith-Jones P, et al. Response to treatment with Yttrium 90-DOTA-Lanreotide of a patient

17. 18.

19.

20.

21.

22.

23. 24.

25. 26. 27. 28.

29. 30. 31.

32.

33.

with metastatic gastrinoma. J Nucl Med 1998;39:2090 – 2094. Wiseman GA, Kvols LK. Therapy of neuroendocrine tumors with radiolabeled MIBG and somatostatin analogues. Semin Nucl Med 1995;25(3):272–278. Forssell-Aronsson E, Lanhede B, Fja¨lling M, et al. Pharmacokinetics and dosimetry of 111In-DTPA-D-Phe1-octreotide in patients with neuroendocrine tumors. In Sixth International Radiopharmaceutical Dosimetry Symposium, Proceedings of a Conference held at Gatlinburg, Tennessee, May 7–10, 1996 (Edited by A Schlafke-Stelson and EE Watson).1999. Cremonesi M, Ferrari M, Zoboli S, et al. Biokinetics and dosimetry in patients administrated with 111In-DOTA-Tyr3octreotide: Implication for internal radiotherapy with 90YDOTATOC. Eur J Nucl Med 1999;26:877– 886. Virgolini I, Szilvasi I, Kurtaran A, et al. Indium-111DOTA-Lanreotide: Biodistribution, Safety and radiation absorbed dose in tumor patients. J Nucl Med 1998;39: 1928 –1936. Andersson P, Forssell-Aronsson E, Gre´tarsdo´ttir J, et al. Biokinetics and dosimetry after repeated injections of 111 In-DTPA-D-Phe1-octreotide. In: Schlafke-Stelson A, Watson, EE, editors. Proceedings of the Sixth International Radiopharmaceutical Dosimetry Symposium. Gatlinburg, TN: May 7–10, 1996 . 1999. Ko¨lby L, Ahlman H, Wa¨ngberg B, et al. A transplanted human midgut carcinoid as a model for somatostatin receptor- and amine transporter mediated radionuclide uptake. Submitted. Bernhardt P, Ko¨lby L, Johansson V, et al. Biokinetics of In-DTPA-D-Phe-octreotide in nude mice transplanted with a human carcinoid tumor. Submitted. Krenning EP, Bakker WH, Kooij PPM, et al. Somatostatin receptor scintigraphy with Indium-111-DTPA-D-Phe-1-octreotide in man: Metabolism, dosimetry and comparison with iodine-123-Tyr-3-octreotide. J Nucl Med 1992;33: 652– 658. Brownell GL, Ellett WH, Reddy AR. Absorbed fractions for photon dosimetry. MIRD pamphlet No. 3. J Nucl Med 1968;1(Suppl. 1):27–39. Ellett WH, Humes RM. Absorbed fractions for small volumes containing photon-emitting radioactivity. MIRD Pamphlet No.8. J Nucl Med 1971;12(Suppl. 5):25–32. Chu SYF, Ekstro¨m LP, Firestone RB. The Lund/LBNL Nuclear Data Search: version 2.0, February 1999. http:// nucleardata.nuclear.lu.se/nucleardata/toi/index.asp Howell RW, Rao DV, Haydock C. Dosimetry techniques for therapeutic applications of incorporated radionuclides. In The joint symposium on dosimetry of administered radionuclides, Proceedings held at Washington D.C, September 1989:215–256. Cole A. Absorption of 20 eV to 50000 eV electron beams in air and plastic. Radiat Res 1969;38:7–33. Muthuswamy MS, Roberson PL, Buchsbaum DJ. A mouse bone marrow dosimetry model. J Nucl Med 1998;39:1243– 1247. Reubi JC, Scha¨r J-C, Waser B, et al. Affinity profiles for human somatostatin receptor subtypes SST1-SST5 of somatostatin radiotracers selected for scintigraphic and radiotherapeutic use. Eur J Nucl Med 200;27:273–282. O’Donoghue JA, Bardies M, Wheldon TE. Relationships between tumor size and curability for uniformly targeted therapy with beta-emitting radionuclides. J Nucl Med 1995; 36:1902–1909. Forssell-Aronsson E, Fja¨lling M, Nilsson O, Tisell LE,

524

I. J. Radiation Oncology

●

Biology

●

Physics

Wa¨ngberg B, Ahlman H. 111In activity concentration in human tissue samples after i.v. injection of 111In-DTPAPhe1-octreotide. J Nucl Med 1995;36:7–12. 34. Sastry KS. Biological effects of the Auger emitter iodine125: A review. Report No. 1 of AAPM Nuclear Medicine Task Group No. 6. Med Phys 1992;19(6):1361–1370. 35. Andersson P, Forssell-Aronsson E, Johanson V, et al. In-

Volume 51, Number 2, 2001

ternalization of 111In into human neuroendocrine tumor cells after incubation with 111In-DTPA-D-Phe1-octreotide. J Nucl Med 1996;37:2002–2006. 36. de Jong M, Breeman WAP, Bernad BF, et al. [DOTA,Tyr3]octreotate, labeled with 177Lu, is most promising for radionuclide therapy of somatostatin receptorpositive tumours. Nucl Med Commun 2000;21(6):569.