1α,25-Dihydroxyvitamin D3 (calcitriol) and its analogue, 19-nor-1α,25(OH)2D2, potentiate the effects of ionising radiation on human prostate cancer cells

Radiotherapy with external beam radiation or brachytherapy is an established therapeutic modality for prostate cancer. Approximately 30% of patients with localised prostate cancer relapse at the irradiated site. Secondary effects of ionising radiation (IR), for example, bowel and bladder complications, are common. Thus, the search for biological response modifiers that could potentiate the therapeutic effects of radiation and limit the occurrence of serious side effects is an important task in prostate cancer therapy. 1α,25-Dihydroxyvitamin D3 (calcitriol), the active metabolite of vitamin D, and its analogues are under investigation for the treatment of several malignancies including prostate cancer. Here, we report that 1α,25-dihydroxyvitamin D3 and its less calcaemic analogue 19-nor-1α,25-(OH)2D2 (Zemplar®) act synergistically with IR to inhibit the growth of the human prostate cancer cells in vitro. 1α,25-dihydroxyvitamin D3 potentiated IR-induced apoptosis of LNCaP cells, and nanomolar doses of 1α,25-dihydroxyvitamin D3 and 19-nor-1α,25-(OH)2D2 showed synergistic inhibition of growth of LNCaP cells at radiobiologically relevant doses of IR (1–2 Gy). At higher doses of IR, the combination of 1α,25-dihydroxyvitamin D3 and IR or 19-nor-1α,25-(OH)2D2 and IR resulted in moderate antagonism. The synergistic effect at radiobiologically relevant doses of radiation suggests that a combination of 1α,25-dihydroxyvitamin D3 or 19-nor-1α,25-(OH)2D2 with IR could permit a reduction in the dose of radiation given clinically and thus potentially reduce treatment-related morbidity.

Adenocarcinoma of the prostate is the most commonly diagnosed nonskin cancer and the second leading cause of cancer death in men in the United States (Keely and Gomella, 1998;Garnik and Fair, 1998;Jemal et al, 2003). Most men will present with clinically localised disease but a significant minority will present with locally advanced disease. In patients with locally advanced disease (T3 -4) the rates of local control with radiation therapy alone have been reported to range from 60 to 77% (Shipley et al, 1995). In a recent Phase III trial in men with locally advanced prostate cancer, the rate of local regional control in the radiation-alone arm was 83% (Bolla et al, 2002). Since this trial did not require prostate biopsies following treatment, the rate of loco-regional control is likely an overestimate.
It is clear that in men with locally advanced disease, optimal local control with radiation therapy has not been achieved. There is some evidence that higher radiation doses may improve local control, but higher doses lead to increased morbidity, especially rectal bleeding (Pollack et al, 2002). Whether used for cure or palliation, side effects of RT are common and include adverse effects on urinary, bowel and sexual function (Hamilton et al, 2001). Evidence also exists suggesting that the cells that survive the initial RT are the ones most likely to repopulate the irradiated area and metastasise (Fuks et al, 1991). Thus, agents or methods that could potentiate the effects of ionising radiation would be very desirable. Ideally, such agents would permit a reduction in the dose of radiation administered and thereby, a potential reduction in the incidence and/or severity of side effects.
The steroid hormone 1a,25-dihydroxyvitamin D 3 (1a,25(OH) 2 D 3 ) , also known as calcitriol, the active metabolite of vitamin D, and its synthetic analogues are currently under intensive investigation for several malignancies, including prostate cancer . Based on the observation that the descriptive epidemiology of prostate cancer resembles that of vitamin D insufficiency, Schwartz and Hulka (1990) proposed that vitamin D maintained the differentiated phenotype of prostate cells and that low levels of vitamin D increase the risk for prostate cancer. Many subsequent epidemiologic (Ahonen et al, 2000;Luscombe et al, 2001) and laboratory studies (Feldman et al, 2000) have supported this hypothesis. For example, 1a,25(OH) 2 D 3 and its analogues exhibit significant growth-inhibitory, proapoptotic, anti-invasive and antimetastatic effects on prostate tumour cell lines in vitro and in vivo (Miller et al, 1992;Peehl et al, 1994;Schwartz et al, 1995Schwartz et al, , 1997Skowronski et al, 1995;Asou et al, 1998;Lokeshwar et al, 1999;Blutt et al, 2000;Seol et al, 2000). These studies support the potential use of 1a,25(OH) 2 D 3 and its analogues as therapeutic agents for prostate cancer (reviewed in Blutt and Weigel, 1999).
The principal limitation of 1,25(OH) 2 D 3 administration is the risk of hypercalcaemia. Consequently, many synthetic analogues of 1,25(OH) 2 D 3 have been developed with the goal of developing analogues that exhibit similar or enhanced growth-inhibitory and antimetastatic properties with reduced calcaemic effects (Schwartz and Hulka, 1990) (see also Hansen et al (2001) for a review). Prominent vitamin D analogues include EB1089 (Seocalcitol), which is currently in Phase III trials in Europe for hepatocellular carcinoma, 1-a-vitamin D 2 and 19-nor-1a,25-(OH) 2 D 2 (Zemplar), a 1,25(OH) 2 D 3 -analogue that is FDA-approved for the treatment of secondary hyperparathyroidism. Laboratory studies have demonstrated that the antiproliferative effects of 19-nor-1a,25-(OH) 2 D 2 on prostate cancer cells are indistinguishable from those of 1a,25(OH) 2 D 3 (Chen et al, 2000).
In addition to its use as monotherapy, calcitriol and analogues may prove useful in conjunction with other treatment modalities, notably, ionising radiation. Observations that support combining 1,25(OH) 2 D 3 with ionising radiation include the following: (a) both IR and 1,25(OH) 2 D 3 induce apoptosis in prostate carcinoma cells by apparently distinct pathways (Welsh et al, 1995;Billis et al, 1998;James et al, 1998;Pirianov et al, 1999;Ding and Fisher, 2001;McGuire et al, 2001;Sheard, 2001;Liu et al, 2002); (b) other agents that induce cellular differentiation, for example, phenylacetate, platelet-derived growth factor and retinoic acid, are known to enhance the cytotoxic effects of IR (Bill et al, 1992b;Miller et al, 1997;Hoffmann et al, 1999); (c) EB 1089 has been shown to enhance the radiation sensitivity of breast carcinoma cells (Sundaram and Gewirtz, 1999). These findings strongly suggest that 1,25(OH) 2 D 3 may enhance the cytotoxic effects of IR on prostate cancer cells. In this report, we investigated the effects of combined treatments of 1a,25(OH) 2 D 3 and 19-nor-1a,25-(OH) 2 D 2 with IR on the apoptosis and growth of the human prostate cancer cell line LNCaP, as well as on the growth of a human primary prostatic cell strain obtained from patients who had undergone radical prostatectomy.

Cell culture
The LNCaP cell line was purchased from the American Type Culture Collection (Rockville, MD, USA) and maintained in RPMI 1640 supplemented with 10% foetal calf serum (FCS). Primary human epithelial cell cultures from histologically normal prostatic peripheral zones or prostate cancer were obtained from radical prostatectomies performed at Wake Forest University School of Medicine as previously described (Peehl and Stamey, 1986;Peehl et al, 1988;Barreto et al, 2000). Briefly, a small piece of tissue from each specimen is removed and minced. The tissue is digested with collagenase overnight. To remove the collagenase and the majority of the stromal cells, the tissue is rinsed and centrifuged. The collagenase-digested tissue is inoculated into a 60-mm tissue culture dish coated with collagen type I (Collagen Corporation, Palo Alto, CA, USA) and grown in medium PFMR-4A supplemented with growth factors and hormones . Previous studies have demonstrated that prostatic stromal cells do not grow in the serum-free conditions used in this study, yet these conditions maintain the growth and differentiation of prostatic epithelial cells (Peehl and Stamey, 1986;Peehl et al, 1988). The cells that grew out from the tissue were aliquoted and stored in liquid nitrogen. The histology of each specimen was verified by inking and fixing the prostate after dissection and serially sectioning the marked area. The histology of sections immediately adjacent to the area of the dissection was reviewed and classified as histologically normal or cancerous by a trained pathologist. The frozen aliquots were thawed to produce secondary cultures, which were grown in medium MCDB 105 (Sigma, St Louis, MO, USA) supplemented with growth factors and hormones (Peehl and Stamey, 1986).

Irradiation
LNCaP cells were plated in 24-well plates at densities of 5000 cells per well. Primary cells were plated onto 35 mm dishes at a cell density of 1 Â 10 4 dish À1 . At the end of preincubation with 1a,25(OH) 2 D 3 or 19-nor-1a,25(OH) 2 D 2 , cells were irradiated in a 137 Caesium irradiator (Shepherd and Associates) at a dose rate of 472 rad min À1 . Following IR, the cells were returned to the incubator and were allowed to grow for another 12 -14 days, before assessment of viability as described below.

Apoptosis assay
Cells were plated in 60 mm plastic dishes at a density of 2 Â 10 5 cells per plate 24 h prior to treatment. Following treatments, determination of the number of apoptotic cells was assayed. Cells with apoptotic morphology are determined by incubating cells with Bisbenzimide (Hoechst 33342) dye and propidium iodide (PI), both at a concentration of 2 mg ml À1 for 10 min. Cells with apoptotic morphology (fragmented nuclei and PI-positive cells) were counted in each of four fields randomly selected in the dish and expressed as a ratio to total number of cells. Both attached and floating cells were counted. The results represent the average of three independent experiments (7s.e.m.).

Cellular proliferation assays
Two types of cellular proliferation assays were employed. For assessing human prostatic cell strain proliferation, trypan-blue exclusion was used as a marker of viable cells 9 -12 days after treatments. The WFU10Ca strain cells were plated onto 35 mm dishes and treated with 1a,25(OH) 2 D 3 or ethanol, h EtOH). After 24 h later, cells were irradiated as described above. To assess the effects of 1a,25(OH) 2 D 3 or 19-nor-1a,25(OH) 2 D 2 on LNCaP cell proliferation, 5000 cells were plated onto 24-well plates and treated with 1a,25(OH) 2 D 3 or 19-nor-1a,25(OH) 2 D 2 . Cells were irradiated 24 h later. After 10 -12 days, cell viability was assessed by the MTT assay according to the manufacturer's protocol. A volume of 200 ml aliquots from each reaction was transferred into a 96-well plate and absorbance values were measured with an automated plate reader (Molecular Dynamics, Sunnyvale, CA, USA). Each experiment was performed in triplicate.

Isobologram analysis of combination treatments
The isobologram analysis was performed using the CalcuSyn software (Biosoft, Cambridge, UK). The analysis is based on the median-effect principle and the median effect equation by Chou (1974): f a /f u ¼ (D/D m ) m , where D is the dose of the drug; D m is the median-effect dose; f a is the fraction affected by the dose, f u the fraction unaffected by the dose, f u ¼ 1Àf a ; and m is an exponent signifying the sigmoidy (shape) of the dose -effect plot.
The Combination Index (CI) ( Table 1) for multiple drug combinations is calculated based on the equation and plot by Chou and Talalay: In the simplest form, a CIo1, ¼ 1 and 41 indicate synergism, additive effect and antagonism, respectively. Chou and Talalay (1984) also recommend a scale based on the value of CI and using symbols ranging from þ þ þ þ þ for very strong agonism to ÀÀÀÀÀ for very strong antagonism. The CI values calculated from our experiments follow this scale.

RESULTS
To test whether 1a,25(OH) 2 D 3 affects the apoptotic response of LNCaP cells to IR, cells were treated with vehicle (0.1% EtOH) or with 100 nM 1a,25(OH) 2 D 3 . After 48 h, the cells were either irradiated with 4 Gy IR or were mock-irradiated. Cells with apoptotic morphology (fragmented nuclei and PI-positive cells) were counted. Figure 1 shows photographs from representative populations at Â 100 and Â 250 magnifications. Treatment with 1a,25(OH) 2 D 3 or IR alone did not induce substantial apoptosis of LNCaP cells, as is evident from the absence of cells with fragmented and/or condensed nuclei in this representative field. However, pretreatment with 1a,25(OH) 2 D 3 for 48 h followed by IR resulted in a significant increase in the percentage of cells with apoptotic morphology (indicated by white arrows). Figure 1B shows the cumulative results obtained by counting cells with apoptotic vs normal morphology in four randomly selected fields from three independent experiments. Although 1a,25(OH) 2 D 3 and IR induced only modest levels of apoptosis (13 and 17%, respectively) compared to basal levels (5%), pretreatment of LNCaP cells with 1a,25(OH) 2 D 3 for 48 h prior to IR resulted in 48% apoptosis. These results indicate that pretreatment of LNCaP cells with 1a,25(OH) 2 D 3 prior to IR sensitises these cells to IR-induced apoptosis.
While apoptotic assays can be useful in determining the shortterm effects of agents on cells, they may not reflect the long-term effects of these agents on cellular proliferation. To investigate the effects of 1a,25(OH) 2 D 3 and 19-nor-1a,25(OH) 2 D 2 on long-term proliferation of prostate cells, we treated LNCaP cells with different doses of the hormones and IR, and assessed cell viability 8 days later using the MTT assay. As seen in Figure 2A and B, both agents induced a dose-dependent decrease in the survival of LNCaP cells. An isobologram analysis of the data revealed that at lower doses of drug and IR, a significant but modest synergistic effect is observed ( Figure 3A and B). At increased concentrations of drug (100 nM) and IR (4 Gy) (point 4 in isobolograms), the two agents exert antagonistic effects on proliferation. These results suggest that the combination of low nanomolar doses of 1a,25(OH) 2 D 3 or 19-nor-1a,25(OH) 2 D 2 with low doses of IR may have beneficial effects in vivo for the treatment of prostate cancer.
We also wanted to test the effects of 1a,25(OH) 2 D 3 on the response of primary prostate tumour cells to IR. Previously, it was shown that 1a,25(OH) 2 D 3 inhibited the growth of primary prostatic cells in a dose-dependent manner (Peehl et al, 1994;Barreto et al, 2000). We first tested a range of doses of 1a,25(OH) 2 D 3 for its ability to inhibit cell growth in a variety of normal and cancer-derived cell clones (data not shown). A dose of 1 nM had only a minimal effect on the growth of all strains tested. We next tested the effects of combined treatments of 1 nM 1a,25(OH) 2 D 3 with increasing doses of IR. As shown in Figure 4 for a cancer-derived strain (WFU10Ca), IR treatments caused a dose-dependent decrease in cell proliferation. Pretreatment of these cells with 1 nM 1a,25(OH) 2 D 3 caused a significant reduction in cell proliferation at 1 and 2 Gy, but it failed to further increase the inhibitory effect of 1a,25(OH) 2 D 3 at a dose of 4 Gy. Therefore, similar to the results obtained with LNCaP cells, radiation and 1a,25(OH) 2 D 3 act synergistically to inhibit cell proliferation at low doses of drug and IR.
A desired effect of radiation sensitisers/enhancers is a differential cytotoxic profile between tumour and normal cells. Thus, we also investigated the effects of the combined treatments of 1a,25(OH) 2 D 3 and IR on the proliferative capacity of normal prostatic stromal cells, which are of fibroblastic origin and thus provide a good model for normal tissue injury. These cells were pretreated with various doses of 1a,25(OH) 2 D 3 for 48 h, after which they were irradiated with doses of 1 and 2 Gy. Table 2 shows the results of selected treatment combinations for these primary normal cells and the corresponding data for LNCaP cells. These data show that with the exception of the combined dose of 1 nM 1a,25(OH) 2 D 3 and 2 Gy, which produced the highest levels of synergy for both the normal and LNCaP cells, the stromal cells were substantially more resistant to combined treatments than LNCaP cells. Furthermore, even at the synergistic combination of 1 nM 1a,25(OH) 2 D 3 and 2 Gy, the overall inhibitory effect on  Figure 3. The DRI for each drug:IR combination is a measure of how much (-fold) the dose of a drug or agent (e.g. IR), in synergistic combination may be reduced at a given effect level compared with the dose of each drug alone. The DRI is another mathematical interpretation of the CI, and CI ¼ 1/(DRI) 1 +1/ (DRI) 2 . +++++, very strong agonism; -----very strong antagonism.

DISCUSSION
1a,25(OH) 2 D 3 and its analogues have shown synergistic activities with a variety of chemotherapeutic agents, including cisplatin and doxorubicin. In breast cancer cells, 1a,25(OH) 2 D 3 and EB1089 were reported to enhance the effects of ionising radiation, although the lack of isobologram analysis in that study does not permit an assessment of a synergistic or additive effect (Sundaram and Gewirtz, 1999). Here, we demonstrate for the first time that in a prostate tumour cell line and primary tumour cells, 1a,25(OH) 2 D 3 and 19-nor-1a,25(OH) 2 D 2 act synergistically with IR to inhibit the growth of these tumour cells.
The synergistic effects of these agents with IR are manifested as an increase in apoptotic index of treated LNCaP cells, and as a decreased proliferation in longer-term assays in LNCaP and primary prostate cells. However, since the synergistic effect on apoptosis was observed with high doses of IR and 1a,25(OH) 2 D 3 , a combination that was antagonistic in longer-term assays, our results suggest that apoptosis is probably not involved in any synergistic long-term antiproliferative effects of these agents. One possible explanation for these results is the following: inhibition of proliferation by 1a,25(OH) 2 D 3 and its analogues and IR involves multiple mechanisms, of which apoptosis is only one component. Terminal differentiation (or permanent cell cycle arrest) and mitotic (necrotic) death due to unrepaired DNA strand breaks also contribute substantially to decreased proliferation. It is possible that 1a,25(OH) 2 D 3 and IR induce apoptosis via distinct mechanisms and therefore exhibit synergistic activation of apoptosis, but that differentiation and/or necrosis work by mechanisms that are mutually inhibited at higher drug or IR doses. This result is not surprising, since similar drug combinations (e.g., cisplatin and 1a,25(OH) 2 D 3 ) also appear to produce antagonistic effects at high doses.
Although short-term apoptotic assays do not always demonstrate a good correlation with overall cell survival in in vitro assays, apoptosis may play a role in vivo. For example, increased apoptosis in a tumour as a result of treatment with a cytotoxic agent may contribute to reoxygenation of previously hypoxic tumour cells, and thus contribute to cytotoxicity of subsequent IR treatments (Brown and Wouters, 1999;2001). In LNCaP cells, the increased apoptotic index after a 48 h incubation suggests that pretreatment of cells with 1a,25(OH) 2 D 3 lowers the apoptotic threshold of the cells to the subsequent stress of IR. A potential mechanism for this finding involves downregulation of the antiapoptotic protein bcl-2, which has been reported to occur as a result of treatment with 1a,25(OH) 2 D 3 or its analogues (Welsh et al, 1995;Simboli-Campbell et al, 1997;Narvaez et al, 2001).
If an increase in apoptosis does not play a major role in the long-term antiproliferative synergistic interaction between 1a,25(OH) 2 D 3 and its analogues, what other potential mechanisms may be involved in this process? First, calcitriol and its analogues induce cellular differentiation, a process in which a precursor cell acquires morphological and biochemical characteristics of specialised lineage (Studzinski et al, 1993). In the processes of tumorigenesis, cells reverse this process and become less differentiated (i.e., they dedifferentiate). Cells undergoing differentiation appear to be more sensitive to ionising radiation than Cells were treated with ethanol (control) or 100 nM 1a,25(OH) 2 D 3 for 48 h prior to treatment with 4 Gy IR and apoptotic cells were counted 12 h after IR. The apoptotic cells from four randomly selected fields were counted (between 50 and150 cells field À1 ). Three independent experiments were performed. Error bars represent7s.e. values. cells that have a more undifferentiated phenotype. For instance, using a murine 3T3-T proadipocyte cell line, Bill et al have demonstrated that although both undifferentiated and differentiated cells sustain similar levels of DNA damage when irradiated, undifferentiated cells exhibit higher levels of DNA repair (Bill et al, 1992a, b). Similar mechanisms may explain the increased sensitivity of cells treated with the prodifferentiating agents phenylacetate and phenylbutyrate (Miller et al, 1997). Indeed, calcitriol has been shown in earlier studies to potentiate the cytotoxic effects of chemotherapeutic agents by inducing cell differentiation (Ravid et al, 1999). Second, treatment of the human prostate cells LNCaP and PC-3 with calcitriol results in cell cycle arrest, primarily at the G 1 /S interphase, while a G 2 /M arrest has also been reported (Eisman et al, 1989;Godyn et al, 1994;Zhuang and Burnstein, 1998). As shown by work with synchronised cell cultures, with the exception of the M phase of the cell cycle (where cells in culture are the most radiosensitive), the majority of cell lines show the greatest sensitivity to ionising radiation in late G 1 / early S phase and during the G 2 /M transition (Nias, 1988), and are the most radioresistant during early G 1 and S phases. Therefore, it is reasonable to expect that pretreatment of prostate cancer cells with 1a,25(OH) 2 D 3 and 19-nor-1a,25(OH) 2 D 2 will increase their sensitivity to IR by inducing the entry of the majority of the cells in these two radiosensitive phases. Effects of combined treatments of 1a,25(OH) 2 D 3 and IR on primary prostate cells. Cells were isolated from patients who underwent radical prostatectomy at WFU as described in Materials and Methods. Cells were plated into 35 mm dishes and treated with 1 nM 1a,25(OH) 2 D 3 . After 24 h, cells were irradiated in a caesium irradiator with the indicated doses and returned to 371C for growth for an additional 10 -12 days. At the end of treatment, cells were trypsinised and counted. Control cells received only 0.1% ethanol. Experiments were performed in triplicate. The example shown is from the WFU10Ca, a cancer-derived primary strain. Similar results were obtained with two other cancer-derived and three normal prostate-derived strains.
Regardless of the mechanism(s) of synergy, our data show that 1a,25(OH) 2 D 3 and 19-nor-1a,25(OH) 2 D 2 (which are both currently in clinical trials for the treatment of advanced prostate cancer) exhibit significant levels of synergy with one of the most widely used therapeutic modalities for prostate cancer, IR. Importantly, this potentiation occurs at low doses of both drug and IR, which fall well within the range of clinically relevant doses. For example, radiotherapy protocols for prostate cancer patients routinely employ multiple daily fractions of IR at 1.5 -2 Gy per treatment for a total dose to the prostate between 64 and 72 Gy (Perez, 1997). As shown in our isobolograms, greatest synergism occurs at the IR doses of 2 Gy. Furthermore, the doses of 1a,25(OH) 2 D 3 and 19-nor-1a,25(OH) 2 D 2 that exhibit synergism are in the low nanomolar range. Intracellular levels of 1a,25(OH) 2 D 3 in the prostate have not been characterised. However, it is now clear that these levels likely exceed by several fold the levels of 1a,25(OH) 2 D 3 in the systemic circulation (picomolar levels (Platz et al, 2000)). This is because both normal and cancerous prostate cancer cells express 1-ahydroxylase, which enables these cells to convert the prohormone 25-OHD 3 , which circulates at nanomolar concentrations, into 1a,25(OH) 2 D 3 , intraprostatically (Schwartz et al, 1998). Prostate cells that express 1-a-hydroxylase are growth inhibited by 25-OHD 3 at nanomolar concentrations (Barreto et al, 2000). Therefore, both 1a,25(OH) 2 D 3 and 19-nor-1a,25(OH) 2 D 2 are capable of acting synergistically with IR at physiological doses, at least in vitro. Animal experiments will be required in order to validate these concepts in vivo.
Another conclusion of our study is that at higher doses of 1a,25(OH) 2 D 3 and 19-nor-1a,25(OH) 2 D 2 and IR, antagonism occurs. This finding is in agreement with similar studies using 1a,25(OH) 2 D 3 , cisplatin and carboplatin (Moffatt et al, 1999). Although the mechanism(s) for this antagonistic effect is unknown, it is well established that the synergy or antagonism between two drugs with different modes of action is largely dependent upon the drug concentrations and ratios. If these findings extend to in vivo models, they would suggest that careful dosage and patient monitoring must be employed if a combination of 1a,25(OH) 2 D 3 or 19-nor-1a,25(OH) 2 D 2 with IR is administered clinically.
The experiments with the primary cell cultures indicate that 1 nM 1a,25(OH) 2 D 3 can also sensitise primary prostatic cells (both tumour and normal) to IR treatments. This finding supports the potential use of 1a,25(OH) 2 D 3 and Zemplar as IR-response modifiers in vivo, since these cell strains retain most of the morphological and biochemical features of human prostatic tissue. Conversely, the ability of certain combinations of these agents to sensitise normal prostatic tissue to the effects of IR raises the possibility of increased radiation-induced adverse effects (i.e., prostatitis) in surrounding normal prostatic tissue. However, the data also suggest that at least in vitro, combinations of 1a,25(OH) 2 D 3 and IR exist that might preferentially sensitise tumour but not normal prostate cells. Ultimately, such effective drug combinations will need to be established experimentally in either animal tumour models of prostate cancer before these agents are administered for therapeutic purposes in the clinic. Finally, it is possible that in a clinical setting, potential damage to normal tissue might be mitigated by brachytherapy in which radioactive 'seeds' are delivered locally to the tumour tissue, which would minimise the exposure of surrounding normal tissue to radiation.
The isobologram analysis and DRI of the effects of these agents on LNCaP cells also suggest that it may be possible to decrease the dose of IR and/or 1a,25(OH) 2 D 3 and 19-nor-1a,25(OH) 2 D 2 in combined modalities, without sacrificing efficacy. This finding has important clinical implications because the incidence of morbidity caused by RT increases with radiation dose (Perez, 1997). For instance, the DRI value of 2.4 for IR in the combination of 1a,25(OH) 2 D 3 (1 nM) and IR (2 Gy) indicates that to achieve the same efficacy level in the absence of 1a,25(OH) 2 D 3 , one would have to give 2.4 times higher dose of IR. In other words, the presence of 1a,25(OH) 2 D 3 allows for a 2.4 times lower dose of IR to obtain the same therapeutic effect. Conversely, the 9.5 DRI value for 1a,25(OH) 2 D 3 indicates that at this combination of doses, the presence of IR allows for almost a 10-fold decrease in the level of 1a,25(OH) 2 D 3 to achieve the same therapeutic effect. As noted above, one of the limiting factors of administration of 1a,25(OH) 2 D 3 is hypercalcaemia. Our results show that by combining the two modalities it may be possible to reduce the levels of 1a,25(OH) 2 D 3 or19-nor-1a,25(OH) 2 D 2 without a significant compromise in efficacy, which would minimise the risk of hypercalcaemia.
In summary, we have shown that 1a,25(OH) 2 D 3 and its less calcaemic analogue 19-nor-1a,25(OH) 2 D 2 show synergistic effects when combined with radiation treatments. Both these drugs are approved for clinical use and both are presently under clinical investigation as monotherapy in prostate cancer. These findings have implications for limiting the disability (e.g., rectal and bladder injury) induced by radiation therapy. Future studies will focus on the in vivo interactions between these treatment modalities (using LNCaP xenograft tumour models) for the treatment of prostate cancer.