Abstract
Background: We developed a new method of brachytherapy, termed diffusing alpha-emitters radiation therapy (DaRT), based on the use of intratumoral 224Ra-loaded wires, which release short-lived alpha-emitting atoms by recoil. Here, we examined their ability to destroy and control the development of several human-derived tumors implanted in athymic mice. Materials and Methods: The experiments were performed on athymic mice bearing malignant human-derived tumors including prostate (PC-3), glioblastoma (GBM, U87-MG), colon (HCT15), squamous cell carcinoma (FaDu) and melanoma (C32). One or more 224Ra-loaded wires were inserted into the tumors, and mice were assessed for tumor growth rate and survival. Results: In vivo studies showed that DaRT can effectively destroy the tumors, and in vitro tests confirmed the sensitivity of the studied cells to alpha particles. While the C32 cells were relatively resistant, other tumor types (e.g. HCT15) exhibited sensitivity in both measured aspects. Conclusion: DaRT could potentially be combined with chemotherapy or other treatment modalities to effectively treat non-resectable tumors.
The efficacy of an anticancer treatment, in general, depends on the sensitivity of the treated tumor to the treatment. Specifically in the field of radiotherapy, many factors might determine the response of a given malignant tissue. Tumor site and origin, cell-cycle stage and the oxygenation state of the cell are considered the most influential parameters on the degree of radiosensitivity or radioresistance. Hence, the ability to predict tumor response to radiotherapy may indicate the outcome of the treatment (1-4).
Due to the fact that the vast majority of methods involving ionizing radiation utilize low-linear energy transfer (LET) radiation (x- and gamma-rays, and electrons), the accumulated experimental data are almost entirely related to this form of energy transfer (5, 6). There are significantly fewer data about the effect of high-LET radiation on different malignancies and in particular about the effect of alpha particles on solid tumors.
Alpha radiation is generally more effective against cancer cells than is low-LET radiation, achieving a higher degree of cell killing probability for a given absorbed dose, and forming complex DNA lesions (7, 8). The biological effect of alpha irradiation has the further advantage of being largely insensitive to oxygen deficiency which is a frequent condition in solid tumors that dramatically reduces the efficacy of radiation treatments based on photons and electrons (9).
The most prominent obstacle that has so far prevented the use of alpha radiation from becoming therapeutically efficient against solid tumors is its range in tissue. The difficulty in creating conditions that would allow effective coverage of the entire tumor volume with alpha particles has been considered too large a hindrance (10, 11).
In previous studies, we described a new form of brachytherapy allowing the treatment of solid tumors by alpha particles (12-15). The method, named diffusing alpha-emitter radiation therapy (DaRT), is based on the utilization of implantable radioactive sources which carry small amounts of 224Ra incorporated into their surface. When 224Ra decays it emits into the tumor its short-lived progeny: 220Rn, 216Po and 212Pb, which leave the source by virtue of their recoil energy. The basic claim of DaRT, namely that the short-lived alpha emitters released from the source can lead to cell death over therapeutically significant distances (while the parent 224Ra remains fixed on the source), was demonstrated in three different experimental models.
The need to evaluate the consistency of the DaRT method and its efficacy in other malignancies has brought us to challenge the treatment in additional experimental models of solid tumor. Since the power of alpha particles to destroy tumor cells is not limited to a specific site or origin, it needs to be applied to various types of malignancies in order to strengthen the understanding of its efficacy under different conditions of tumor microenvironments. The focus on human-derived tumors in this work is evidently an essential step in promoting the method towards clinical trials.
Materials and Methods
Cells and tumors. HCT15, FaDu and PC3 cell lines were purchased from the American type culture collection (ATCC) and their use begun within six months. The U87 cell line was kindly provided by Professor Yoel Kloog (Life-Sciences Faculty, Tel-Aviv University). Cells were grown in Dulbecco's Modified Eagle Medium (DMEM) in 5% CO2 at a humidified incubator with 37°C. All media were supplemented with 10% fetal calf serum, sodium pyruvate (1 mM), non-essential amino acids (1%), L-glutamine (2 mM), penicillin (100 U/ml), and streptomycin (100 μg/ml) (all obtained from Biological Industries, Beit-Haemek, Israel). Athymic nude mice (8 to 12 weeks old) were obtained from Harlan (Rehovot, Israel). Animal care and experiments were carried out in accordance with the guidelines of the Israeli National Council for Animal Experimentation (permits no. M-06-015, M-06-070 and M-07-060). All surgical and invasive procedures were conducted under anesthesia by intraperitoneal inoculation of Imalgen (100 mg/kg; Fort Dodge Laboratories, Germantown Hills, IL, USA) and xylazine hydrochloride (10 mg/kg; VMD, Arendonk, Belgium) solution in 0.25 ml of phosphate buffers solution (PBS). Xenografts were transplanted subcutaneously into the low lateral side of the back. Local tumor growth was determined by measuring 3 mutually orthogonal tumor dimensions with a digital caliper (Mitutoyo, Onomy, Japan). The volume of tumor was calculated using the approximate formula: V=π/6 D1D2D3 where D1, D2, and D3 are the measured dimensions.
Colony formation assay. The in vitro effect of alpha particles traversing various cell lines was studied using a 228Th source (half life 1.91 years) in a setup described previously (15). It consists of a small deposit of 228Th in secular equilibrium with its daughters on a silicon substrate. The surface was coated with a 4.0-μm layer of silver to prevent the release of the recoiling daughters of 228Th. The irradiation setup comprised a 7.5-μm thick Kapton (polyimide) foil (Dupont, Manon, Luxembourg) held between two cylindrical stainless steel parts, forming a well with an inner diameter of 9 mm. Cells were placed in the wells and were incubated for 24 h to form a confluent monolayer, after which the wells were exposed to alpha particles for 0, 1, 2, 3, 4 and 6 min, at an average dose rate of approximately 0.5 Gy/min. The dose rate was estimated by a measurement of the total flux of alpha particles crossing the Kapton foil using an alpha particle detector, and a Monte Carlo simulation of the passage of alpha particles through the foil using the ‘stopping and range of ions in matter’ (SRIM) code. Immediately following irradiation, cells were harvested and sparsely seeded in 10-cm culture dishes (Corning, Corning, NY, USA) with 15 ml of medium, to allow the formation of well-separated colonies. Colony formation was monitored daily, and typically 6-9 days post irradiation colonies were fixed in methanol and stained with Hemacolor (Millipore, Dubllin, Ireland). The colonies were counted manually and were considered viable when containing more than 50 cells.
High-resolution autoradiography (HRA). The radionuclide spread inside treated tumors was studied by an autoradiographic method utilizing a phosphor-imaging plate, as described in a previous publication (12). Briefly, tumors were treated with a single 224Ra-loaded wire each and were excised four days later, with the wire extracted within ~10 min. The excised tumors were then placed in 4% formaldehyde for 24 h and the preserved specimens were subsequently processed and embedded in paraffin following standard procedures. Histological sections (10-μm thick) were cut and placed on glass slides. The slides were then laid on a Fuji imaging plate (BAS-TR2040S, Fujifilm, Tokyo, Japan) for 10-15 h, to measure the 212Pb spread throughout the tumor sections. Following each measurement, the plate was scanned by a Fuji FLA-9000 system with a pixel size of 100 μm. The measured 212Pb activity was then used to estimate the absorbed dose and the asymptotic dose was further normalized by the initial rate of 220Rn release from the source. The same histological specimens were stained with hematoxylin-eosin (H&E) (Surgipath, Richmond, IL, USA) for tissue damage detection, to be correlated with the activity distribution measurements.
224Ra-loaded wire preparation and insertion. 224Ra-loaded wires were prepared using a 228Th generator [as described in detail in (12)]. In this set-up, positive 224Ra ions emitted by recoil from a surface layer containing 228Th are collected electrostatically on a thin stainless stee wire (0.3 mm in diameter; Golden Needle, Suzhou, China). The wires are subsequently heat-treated to induce radium diffusion away from the surface to a typical depth of 5-10 nm. The 224Ra-impregnated wires are characterized by an alpha particle detector to account for their 224Ra activity and the release rate of 220Rn. Wires, either loaded with 224Ra or inert, cut to a length of 4-6 mm were placed near the tip of a 21-gauge needle attached to a 2.5 ml syringe (Picindolor, Rome, Italy) and inserted into the tumor by a plunger placed internally along the syringe axis. Typical wire activities were in the range of several dozen kBq 224Ra.
Results
Variable cell-dependent sensitivity to alpha particles. A set of experiments testing cell lines derived from different human solid malignant tissues was performed in order to compare the relative sensitivity of each cell line to measured fluxes of alpha particles. Escalating doses of radiation were delivered through cell-containing Kapton wells and a comparative analysis was conducted.
For each cell line, two colony formation experiments were performed, with three to five wells irradiated in each experiment for each dose level. The surviving cell fraction was calculated as the ratio of the number of viable colonies in a given culture dish (containing irradiated cells) to the average number of colonies in the unexposed control dishes. Survival curves obtained for C32, HCT15, PC3 and FaDu cells were drawn and the data in each experiment were fitted with the function f(D)=e−D/D0 using Matlab's curve fitting tool to estimate the mean lethal dose (D0). The resulting average values for D0 are given in Table I.
Dose distribution inside treated tumors from different human histotypes. The dispersion of radioisotopes released from the 224Ra wire to further emit alpha particles throughout the malignant tissue was evaluated using an HRA system. We examined tumors treated by an intra-tumoral insertion of a single 224Ra wire and compared the dimensions of the region inside the tumors in which an effective therapeutic dose of alpha radiation (10 Gy) was achieved. The choice of 10 Gy as a marker dose was derived from estimations of low-LET effective doses of 20 Gy and considered a relative biological effect (RBE) of two-fold in the case of alpha particles. The following human derived malignancies were tested: C32 melanoma cell line (n=4), FaDu squamous cell carcinoma (SCC) (n=4), colon carcinoma HCT15 (n=5), U87 (n=4), head and neck SCC CAL27 (n=3), and prostate carcinoma PC3 (n=2). Autoradiography indicated that there are significant differences in the distribution pattern between tumors of different histotypes (the findings derived from our analysis are summarized in Table I). The average area of the region subjected to a normalized asymptotic dose of 10 Gy (per 1 μCi of 220Rn released from the source at insertion) was consistent across the tumors of given cell lines, but varied from 9 to 21 mm2 across different cell lines (with the minimum spread detected in the C32 melanoma model and the maximal spread in FaDu and U87) (Figure 1). The corresponding average diameters of the 10 Gy region varied, accordingly, from 3.5 to 5.2 mm.
Tumor growth retardation. Colon: We investigated the effect of 224Ra wires on colonic tumors derived from a human cell line. Randomized athymic mice inoculated with HCT15 cells were treated with either 224Ra loaded or inert wires when tumors reached 5 mm (average lateral diameter). Figure 2A shows a pronounced effect of the radioactive wires on tumor development monitored three times a week. Thirty-eight days post-treatment, the average volume of the 224Ra-treated tumors (n=7) was 4.7 times smaller than that of the inert group (n=8). The photographs shown in Figure 2 were taken 33 days after treatment, emphasizing the difference between a mouse treated with a single 224Ra wire (Figure 2D), in which tumor suppression was observed, as opposed to an untreated mouse (Figure 2E).
Prostate: PC3 tumors were either treated with radioactive wires or inert wires. Twenty-nine days post treatment, the average volume of the 224Ra-treated tumors (n=6) was three times smaller than that of the inert group (n=5) as can be seen in Figure 2B. Furthermore, one case of complete cure with no tumor recurrence was observed in the group treated with the radioactive wires.
Relative sensitivity to alpha particles by histotype.
Glioblastoma: Athymic mice inoculated with human derived U87 tumor cells (4 mm in average diameter) were treated with either 224Ra or inert wires. The results shown in Figure 2C indicate that after 24 days, the average tumor volume of the 224Ra-treated group was 10 times smaller compared to that of the inert group. There was one case of complete regression of the tumor in the 224Ra wire-treated group and in the inert –wire treated group.
Melanoma: C32 human melanoma tumors induced in nude mice were treated at an average lateral diameter of 8 mm with either 224Ra-loaded (n=8) or inert (n=8) wires. An interesting phenomenon was observed, as the radioactive treatment did not significantly affect tumor growth. We noticed that untreated animals died when tumors reached an approximate volume of 1000 mm3. We therefore determined 1000 mm3 as an end-point value for the C32-transplanted mouse experiments. A statistically significant difference (χ2=0.002) between the inert-treated group and the 224Ra wire-treated group was shown as six out of eight animals in the inert group had died when tumors reached 1000 mm3 compared to none in the radioactive treatment group. These results were consistent with an additional experiment utilizing a double wire insertion method (data not shown).
Discussion
This study examined several types of solid human tumors differing in site for the treatment with alpha-emitting atoms released interstitially. The choice of malignancies took into consideration sites, which may potentially become clinical candidates for this radiotherapy method. Effective treatment of GBM is complicated by multiple factors, including the difficulty in overcoming the blood-brain barrier and the diffusely infiltrative nature of the disease, which limits complete surgical resection (16, 17). Brachytherapy has been demonstrated in clinical trials to be effective for relapsed GBM and the concept of using alpha radiation may give rise to a new clinical tool. For prostate cancer, where brachytherapy is already well-established but still not free of complications, a potential advantage of using DaRT (in conjunction with other methods) may be the ability to deliver a highly localized dose to the region surrounding the urethra while sparing the urethra itself (18, 19). The standard treatment for colonic carcinoma and melanoma, on the other hand, does not include any interstitial irradiation at present. Nevertheless, the relative accessibility of these tumors, together with the fact that surgical procedures fail to yield satisfying cure rates for advanced stages of these diseases, have brought us to investigate the response of these tumors to DaRT application.
Normalized asymptotic dose formed by interstitial 224Ra wires in different human solid tumors. High-resolution autoradiography (HRA) analysis of tumors from different histotypes was performed in order to evaluate the average spread of 224Ra progeny (namely 212Pb) inside the malignant tissue. Average values measured in all tested models are presented in Table I. Representative distributions subsequently stained with H&E to correlate with necrotic domains are presented from HCT15 (A), U87 (B) and C32 (C) models.
Human colonic (A) prostatic (B) or glioblastoma tumors (C) treated by a single 224Ra wire applied to athymic mice bearing tumors and monitored for tumor growth. Control group compared to tumor-bearing mice treated with inert wires. A: HCT15: tumor sizes (average diameter) on treatment day were 5 mm. The 224Ra wire carried activities ranging from 20-24.2 kBq. Inert n=8; 224Ra wire n=7. B: PC3: tumor sizes (average diameter) on treatment day were 3 mm. The 224Ra wire carried activities ranging from 15.3-25.4 kBq. Inert n=5; 224Ra wire n=6. C: U87: tumor sizes (average diameter) on treatment day were 4 mm. The 224Ra wire carried activities ranging from 21.3-24.3 kBq. Inert n=7; 224Ra wire n=7. Standard deviations are represented by bars. Representative photographs of Inert-treated animal (E) and 224Ra wire-treated animal (D) bearing HCT15 tumors 33 days post-treatment are displayed.
Human melanoma tumors treated by a single 224Ra wire-treatment applied to athymic mice bearing tumors and monitored for tumor growth. 224Ra wire (n=8). Tumor-bearing mice treated with 224Ra wires. The control group was referred to as Inert (n=8) tumor-bearing mice treated with inert wires.
We report here that when wires impregnated with small amounts of 224Ra (in the range of a few dozens of kBq) are inserted into different solid human tumors, they may lead to considerable local tumor growth inhibition. The efficacy of the treatment varies, to some extent, across cell lines, and might be affected by several factors including, in particular, the effective diffusion rate of the radionuclides released from the source in each histotype and the relative sensitivity at the cellular level.
When alpha particles traversed cell monolayers in vitro, a scope of responses ranged between the relatively resistant cell lines (C32) to the most alpha radiation sensitive cells (FaDu). The differences, which were significant, implied that the effect caused by this high LET radiation is cellular dependent. The further comparison between histotypes, in regard of the formation of an effective dose region inside the treated tumor, gave rise to yet another range of values. These two tested elements may suggest diversity in the responses to DaRT at two separate levels, so that if the cells inside a bulky tumor require larger doses in order to be eradicated and additionally the spread of the atoms inside the tissue is relatively small, the tumor will only be moderately affected and vice versa. The assumption about the linkage between cell sensitivity and spread must be further examined since it may hold the key for understanding what controls the basic principles of this technique. As already demonstrated both in vitro and by histological means, the effect of DaRT on primary tumors differs across cell lines. Here, again, the results from the in vitro cytotoxicity experiments and HRA experiments were mostly correlative with the effects on tumor volume. The most prominent example is the C32 melanoma model, where we observed that cells were alpha radiation-resistant and the perimeter of the effective dose was relatively small and, indeed, tumor progression was only moderately reduced (Figure 3). On the other hand, it was demonstrated that U87 tumors could be sufficiently controlled by the interstitial 224Ra wires as the radioactive treatment yielded a reduction of over 90% in tumor volume compared to the control animals at 24 days post- wire insertion (Figure 2C). Both the colonic carcinoma model and the prostate carcinoma model were also affected significantly by the intra-tumoral irradiation giving rise to growth arrest of a 4-fold decrease for the PC3 tumors (Figure 2B) to a 4.7-fold decrease for the HCT15 tumors (Figure 2A). In order to rule out any additive damaging effects of non-alpha progeny in this decay chain (i.e. electrons and photons), we used sealed DaRT wires ensuring nothing but alpha-emitting daughters infiltrated the tumor (data not shown). Despite the fact that C32 melanoma primary tumors did not respond to the treatment, the phenomenon in which animals treated with inert wires were highly affected by tumor size, while DaRT wire-treated animals survived, indicated an involvement in the metastatic process (which eventually leads to morbidity). The manner by which the interstitial irradiation using alpha-emitting atoms may promote anti-metastatic processes should be thoroughly investigated.
The probability of eradicating a single cultured cell with alpha particles may vary and be dependent on several characteristics such as the ability to initiate DNA repair mechanisms and recover from complex double-strand breaks (20, 21). Inducing cell death would be even more complicated when the irradiation is aimed against cells dwelling inside a bulky solid tumor due to the fact that the tumor microenvironment becomes a key factor in determining radiosensitivity and a large diversity may be expected between malignancies not sharing the same kind of vasculature, stroma or density for instance (22, 23).
Intensive research efforts must be dedicated to recognizing the factors tilting the balance from radioresistance to radiosensitivity when an alpha particle intercepts a given tumor cell in situ. The pre-clinical results presented here may in the future help predict if DaRT will serve as a good practice option for a given tumor.
Acknowledgements
This study was funded in part by Althera Medical, Ltd. (Tel Aviv, Israel), and by a grant from The Israel Cancer Association. TC, MS, LA, IK and YK. received financial support from Althera Medical Ltd. The research was carried out in the Tel Aviv University. Althera Medical Ltd. asserted no control over the Author's right to publish the results.
Footnotes
-
Conflict of Interests
The Authors declare conflict of interests as detailed in the acknowledgment section.
- Received September 4, 2012.
- Revision received October 18, 2012.
- Accepted October 19, 2012.
- Copyright© 2012 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved