Calcitriol is the principal biologically active metabolite of vitamin D. Calcitriol's activity against many neoplasms is well documented, but calcitriol's therapeutic application has been hampered by predictable hypercalcemia when it is given daily. Because laboratory data has suggested that intermittent exposure to high levels of calcitriol may be sufficient to produce antiproliferative effects, the authors developed a Phase I trial to determine the maximal tolerated dose, dose-limiting toxicity, and the pharmacokinetic profile of calcitriol given weekly by mouth.
Patients with refractory malignancies were enrolled for 4 weeks of treatment followed by 4 weeks of observation. Reenrollment at a higher dose level was permitted for patients who had evidence of response or stable disease and no Grade 3 or greater toxicity. The starting dose was 0.06 μg/kg.
Fifteen patients received 20 cycles of therapy. Doses up to 2.8 μg/kg of calcitriol weekly produced no dose-limiting toxicity. While peak levels and the area under the serum concentration-time curve of calcitriol increased in a linear fashion at lower doses, saturable absorption was observed at doses above 0.48 μg/kg. Doses of 0.48 μg/kg and above produced mean peak calcitriol levels of 1625 pg/mL, approximately 25-fold greater than top normal levels and well within the therapeutic range suggested by in vitro experiments. Eight patients experienced self-limiting Grade 1 hypercalcemia.
Calcitriol (1α,25-dihydroxycholecalciferol; 1,25-dihydroxyvitamin D3), the principal biologically active form of vitamin D, exerts its activity through binding the nuclear vitamin D receptor (VDR) and through less well-characterized nongenomic pathways. In addition to maintaining calcium homeostasis, vitamin D is a mediator of proliferation, differentiation, and immunoregulation.1
In vitro assays using calcitriol or its analogs demonstrate antiproliferative effects in cell lines derived from carcinomas of the prostate,2–4 breast,5 lung,6 colon,7 pancreas,8 and endometrium,9 myeloid leukemia,10 melanoma,11 and sarcomas of the soft tissues12 and bone.13 Studies in animals, although often limited by hypercalcemia, show antiproliferative activity of calcitriol or its analogs in prostate,4 breast,14 and squamous cell carcinomas,15 myeloid leukemia,10 and retinoblastoma.16
In vitro and in vivo experiments have suggested a variety of possible mechanisms for calcitriol's antiproliferative activity including alterations in cell cycle control proteins, growth factor signaling, metalloproteinase activity, angiogenesis, and expression of apoptosis inhibitors. Calcitriol induces cell cycle arrest in G1,17 which may be mediated by induction of the cyclin-dependent kinase inhibitors p21(waf1) and p27(kip1).18, 19 It leads to dephosphorylation of the retinoblastoma protein20 and down-regulates Bcl-2 expression.21 Calcitriol decreases the expression of epidermal growth factor receptors,22 induces transforming growth factors TGFβ1 and/or TGFβ2,23, 24 and alters levels of an insulin-like growth factor IGF-1 and several IGF-binding proteins.25, 26 Calcitriol also inhibits prostate carcinoma invasiveness. This effect may occur through decreases in the activity of type IV collagenases MMP-2 and MMP-9.27 In animal models, calcitriol inhibits angiogenesis.28
Calcitriol, as the metabolically active form of vitamin D3, regulates calcium homeostasis. Antiproliferative effects occur at supraphysiologic calcitriol levels, which predictably cause hypercalcemia when calcitriol is administered daily. Prolonged treatment with doses just above replacement produced little effect in advanced prostate carcinoma,29 myelodysplastic syndrome,30 and ovarian carcinoma.31 However, in a study that had enrolled patients with a rising prostate-specific antigen (PSA) level after they had received definitive treatment for localized prostate carcinoma, daily calcitriol reduced the slope of the rise in PSA.32
Significant dose escalation was possible when calcitriol was administered subcutaneously every other day. At this dose, the maximal tolerated dose (MTD) was 10 μg, and achieved peak levels reached 288 pg/mL.33
Pulse dosing may unlink calcitriol's beneficial and toxic effects. In tissue culture, prostate carcinoma growth inhibition requires calcitriol levels of 1.5–10 times the normal levels in human blood,2, 34 and intermittent exposure to calcitriol may be sufficient for sustained antiproliferative effect.2 Further, pulse dosing may be particularly important when calcitriol is combined with cytotoxic agents. In the squamous cell carcinoma model, growth arrest is observed in vitro after 24 hours of treatment.35 Twenty-four hour treatment also is sufficient to induce retinoblastoma protein dephosphorylation, decreased expression of p21(Waf2/Cip1), and increased expression of p27(Kip1).36 Indeed, sensitization to chemotherapy by calcitriol or its analogs has been demonstrated in mouse model and in vitro systems of squamous cell and prostate carcinomas.35, 37
These data, in conjunction with pharmacokinetic and toxicity data in normal volunteers and patients with osteoporosis,38–40 led us to the hypothesis that weekly dosing of calcitriol is likely to achieve serum levels sufficient for a biologic anticancer effect without causing osteoclast activation or significant hypercalcemia. Based on a finding that mice tolerate higher doses of calcitriol when placed on a calcium restricted diet,14 we hypothesized further that dietary control of calcium may reduce the likelihood of adverse events. The weekly schedule was chosen in part because of its suitability for future combinations with weekly chemotherapy.
MATERIALS AND METHODS
Eligibility criteria included histologically confirmed malignancy refractory to standard therapy, age ≥ 18 years, expected survival of > 2 months, Eastern Cooperative Oncology Group performance status ≤ 2, hematocrit ≥ 30, serum creatinine ≤ 1.2 mg/dL, serum calcium ≤ 10.5 mg/dL, serum phosphate ≤ 4.2 mg/dL, alanine aminotransferase (ALT) ≤ 60 IU/L, total serum bilirubin < 2 mg/dL. Exclusion criteria included pregnancy, history of hypercalcemia, kidney stones, heart failure, or significant heart disease including myocardial infarction in the last 3 months, known cardiac ejection fraction of < 30%, current digoxin therapy, thiazide diuretic therapy within 7 days, bisphosphonate treatment within 4 weeks, systemic steroid therapy within 2 weeks, and unwillingness or inability to stop all magnesium-containing antacids, bile resin-binding drugs, or calcium supplements for the duration of the study. The protocol was approved by the Institutional Review Board. Written informed consent was obtained from each patient before any treatment was started.
Patients were asked to maintain a reduced calcium diet for the 4 weeks of treatment. The goal of the diet was to limit daily calcium intake to < 500 mg. This goal was to be accomplished by limiting the intake of high-calcium foods, including dairy products, foods with milk or milk products as a main ingredient, and high-calcium vegetables, legumes, and seafood. Patients also were instructed to drink 4–6 cups of water in addition to their usual fluid intake beginning 12 hours before each dose and continuing for 3 days. Diet instructions were given by a dietitian before treatment and reinforced during each weekly treatment visit.
Calcitriol (Rocaltrol, a registered trademark of Roche Pharmaceuticals) was given orally once a week for 4 weeks. Calcitriol was formulated in gelatin capsules containing 0.5 μg of calcitriol suspended in butylated hydroxyanisole, butyrated hydroxytoluene, and fractionated triglyceride of coconut oil. Patients were then followed for an additional 4 weeks for toxicity. At the end of 8 weeks, patients who did not have tumor progression or Grade 3 or greater toxicity were permitted to reenroll and to receive either the same dose or the next higher dose of calcitriol. Each weekly dose was given in 4 divided doses given hourly over 4 hours. It was anticipated that even a modest escalation of calcitriol dosing resulting from the pulse dosing would require that a large number of capsules be ingested. Dosing over 4 hours was chosen to reduce the burden of swallowing a large number of capsules at one time. The starting dose was 0.06 μg/kg.
Baseline evaluation included a complete history and physical examination, complete blood count, measurement of levels of serum creatinine, serum calcium, serum phosphate, total serum bilirubin, ALT, alkaline phosphatase, and albumin, serum beta human chorionic gonadotropin (βhCG) level in women of childbearing potential, 24-hour urine collection for calcium, and tumor measurements.
The complete blood count, and serum creatinine, total serum bilirubin, ALT, and alkaline phosphatase levels were monitored weekly. Serum calcium and phosphate levels were checked on the treatment day (Day 1), and on Days 2 and 3. A 24-hour urinary calcium excretion was measured on Day 2.
Compliance with the diet was monitored with a 7-day quantitative food frequency questionnaire directed at calcium-rich foods. Daily calcium intake was estimated by adding the calcium content of calcium-rich foods identified by the questionnaire to the estimated calcium content of the basal diet. The calcium content of the basal diet was calculated to be 1 mg of calcium/8 Kcal. Caloric intake was estimated with the use of the Food Processor 7.0 software (ESHA Research, Salem, OR). The program estimates the daily caloric need based on age, gender, weight, and activity level. Dietary questionnaires that include a comprehensive listing of foods that contribute the relevant nutrient, such as the one used in this study, correlate with more verifiable methods for indicating dietary intake.41 Nevertheless, the method used for this study provided only an estimate of dietary calcium intake.
After completing the 4-week treatment period, patients were monitored for 4 additional weeks. Serum calcium level was checked in Weeks 5 and 7, and tumor measurements were obtained in Week 7. All toxicities were graded according to the National Cancer Institute Common Toxicity Criteria. Response was assessed according to the World Health Organization guidelines.
The planned dose escalation was governed by the multistage escalation scheme described by Gordon and Willson.42 Until the first Grade 3 toxicity was encountered, the accrual of 1 patient was required before dose escalation. Three patients at each dose level were required after the first Grade 3 toxicity. Six patients at each dose level were required when more than 1 Grade 3 toxicity occurred at any dose level. Each subsequent dose level was no more than double the previous level until the first Grade 3 toxicity. A maximum increment of 1.33× over the preceding level was planned after the first Grade 3 toxicity. The MTD was defined as the dose at which ≤ 33.33% of the patients experienced Grade 3 toxicity.43
Patients who had evidence of response or stable disease and no Grade 3 or greater toxicity were permitted to reenroll and receive either the same dose or the next higher dose of calcitriol.
A detailed time course of plasma calcitriol levels was obtained on the initial week of therapy for two of the patients. Timed samples of peripheral plasma were obtained for 18 of 20 patients at time zero, 6 hours, 24 hours, and 48 hours after administration of the Week 1 dose of calcitriol. 1,25-dihydroxyvitamin D levels were measured by a radioreceptor assay using calf thymus 1,25-dihydroxyvitamin D receptor according to the method of Reinhardt, et al.44 Water blanks, control pools, and a standard were processed with the samples to assess accuracy and precision. The AUC was calculated using the linear trapezoidal method.45
Fifteen different patients (Table 1) were enrolled in 20 cycles of therapy. At certain dose levels (i.e., 0.48, 1.60, 2.0, and 2.8) additional patients beyond the minimum required by the dose escalation were enrolled to gain additional pharmacologic data. In Patient 6, the dose escalation was attenuated from the permitted maximum of 2× to 1.66×. Because this patient weighed 1.65× the weight of the patient who had completed the previous dose level, a 2× dose escalation in the weight-based dose of calcitriol would have resulted in a 3.3× escalation of the actual dose. The more conservative multiple was chosen to protect patient safety in this study that already had a relatively aggressive dose escalation scheme. Subsequent patients were enrolled at 2× the highest dose completed at the time of their enrollment. This approach was consistent with the dose escalation scheme that defined the minimum number of patients and dose levels required and the maximum dose escalation between dose levels.
Table 1. Characteristics of Study Patients
No. of patients
ECOG: Eastern Cooperative Oncology Group.
ECOG performance status
Adenocarcinoma of the prostate
Adenocarcinoma of the breast
Adenocarcinoma of the lung
Squamous cell carcinoma of the tonsil
Gastrointestinal stromal tumor
Adenocarcinoma of the pancreas
Bidirectionally measurable disease
Dose escalation was carried out until the dose of 2.8 μg/kg dose was reached without dose-limiting toxicity. Because plasma calcitriol monitoring suggested a plateau in both plasma levels and AUC beyond 0.5 μg/kg, further dose escalation was not attempted, and 6 patients were enrolled at 2.0 μg/kg to gather adequate safety data at doses higher than those planned in future Phase II studies.
Two patients were withdrawn from the study before they completed the 4-week regimen. Both were withdrawn after the third week because of disease progression. No patient withdrew because of toxicity of therapy or unacceptability of the diet. Five of the nine patients who had stable disease opted to reenroll for a second cycle of treatment.
No deaths occurred. No patient withdrew from the study because of toxicity. No Grade 3 or higher toxicity was seen. All observed toxicities (Table 2) were self-limited. All patients who experienced Grade 2 laboratory abnormalities (anemia, leukopenia, and alkaline phosphatase elevation) had Grade 1 abnormalities before they entered the study. In no case did any Grade 1 or Grade 2 toxicity necessitate intervention. Eight of the courses were associated with transient, mild (Grade 1) elevations of serum calcium. Therapy on subsequent weeks was not withheld for Grade 1 hypercalcemia, and no patient developed progressive hypercalcemia of Grade 2 or higher.
Table 2. Toxicities Developed during Each Treatment Course (n = 20)
Grade 3 or 4
All were within normal limits of our laboratory (3.4–10.0 k/mm3) but fell into the Grade 1 toxicity range of 3.0–3.9 k/mm3.
All had Grade 1 abnormalities before entry into study.
Blood samples were obtained from patients to determine the pharmacokinetics of calcitriol administered on this dose and schedule. Calcitriol is a normal component of plasma, and all patients had detectable calcitriol levels in plasma obtained at baseline. The mean baseline levels for plasma calcitriol in this group of patients was 0.16 nmol (range, 0.05–0.28 nmol). The reported normal range for the assay used is 0.05–0.16 nmol. Nine patients had baseline calcitriol levels modestly above the normal range (Table 3).
Table 3. Pharmacokinetic Modeling for Each Treatment Course (n = 18)
Pretreatment calcitriol level (pg/mL)
Net AUC 0–48 hr (ng/mL/hr)
Peak calcitriol concentration observed (pg/mL)
Mean: 68.6; Range: 21–118.
A detailed time course of calcitriol levels was obtained on the initial week of therapy for 2 of the patients treated at the 1.60 μg/kg dose level (Figure 1). Calcitriol levels rose dramatically above baseline, and peak levels of 45-fold and 63-fold above baseline were observed at 4 hours and 8 hours following initiation of ingestion, respectively. Subsequently, the time required for a 50% reduction from peak increment of calcitriol was between 5.6 hours and 6.3 hours. This was very similar to the 5- and 8-hour initial half-time disappearances of orally administered calcitriol when 4 μg doses were given to normal volunteers.38 By 72 hours, calcitriol levels had returned to the range of baseline levels. The overall exposure or AUC of these 2 patients to calcitriol for the 48 hours following ingestion were 22.6 and 26.1 ng/mL hours. These represent 11-fold and 20-fold elevations over the untreated exposure to endogenous calcitriol calculated using the baseline levels of calcitriol.
For these patients, the estimated AUC0–48 hours values were very similar (± 15%) whether all the time points were used or only the 0-, 6-, 24-, and 48-hour values were used for the calculations. Thus, the more limited time course of plasma calcitriol levels taken at 0, 6, 24, and 48 hours that were available from 18 of the 20 initial courses of treatment were used to estimate AUC0–48 hours. The AUC due to calcitriol administration (net AUC) was determined by subtracting the baseline value of calcitriol from each time point. These values are shown in Table 3, and a plot of net AUC versus dose is shown in Figure 2.
The AUC due to calcitriol administration rose through the dose at a range of 0.06–0.24 μg/kg but plateaued at 0.48 μg/kg and beyond. Because the chosen dose escalation scheme favored rapid dose escalation in the absence of serious toxicity, relatively few patients were enrolled at the lower dose levels. No further increase in either the observed peak (6-hour sample) calcitriol concentration or the AUC was observed for doses as high as 2.8 μg/kg. The peak plasma calcitriol levels observed (6-hour samples) rose in a similar fashion and reached a plateau (range, 1500–2784 pg/mL) at the 0.48 μg/kg dose level despite continued escalation of calcitriol as high as 2.8 μg/kg.
Effects on Calcium Metabolism
Mean serum calcium (normal range, 8.5–10.5 mg/dL) increased from 9.55 (standard deviation [SD], 0.57) mg/dL before treatment to 9.76 (SD, 0.63) mg/dL 24 hours later and to 9.88 (SD, 0.68) mg/dL at 48 hours (P = 0.0002 by a two-way, repeated-measures analysis of variance). All calcium levels above the normal range returned to normal within 2 days with no intervention. Mean serum phosphate (normal range, 2.2–4.2 mg/dL) increased from 3.43 (SD, 0.56) mg/dL before treatment to 3.98 (SD, 0.57) mg/dL 24 hours later and dropped to 3.86 (SD, 0.53) mg/dL at 48 hours (P < 0.0001 by a two-way, repeated-measures analysis of variance). Mean 24-hour urinary calcium excretion (normal range, 100–300 mg) increased from 130 (SD, 62) mg with a range of 44–292 mg before treatment to 263 (SD, 125) mg with a range of 59–594 on treatment, measured on Day 2 of each treatment week (P < 0.0001 by a one-way, repeated-measures analysis of variance). There were no statistically significant week to week differences in urinary calcium excretion during treatment. The mean increase in 24-hour urinary calcium excretion was 196 (SD, 162) mg in patients treated with 0.48 mcg/kg and 193 (SD, 80) mg in patients treated at doses of 2.0 μg/kg or higher. These increases were not statistically different (P = 0.97 by Student t test). The increases in urinary calcium excretion were similar in magnitude to urinary calcium increases tolerated indefinitely by patients with asymptomatic hyperparathyroidism, although these patients have a different temporal pattern of development and resolution.46
The primary goals of the dietary intervention in this study were to improve patient safety and to reduce the possibility of hypercalcemia. The dietary intervention was well tolerated and appeared effective at decreasing reported oral calcium intake. No patient withdrew from the study because of the diet. Before study entry, 14 of 20 patients were consuming in excess of 500 mg of calcium daily. The estimated mean daily calcium intake decreased from 824 (SD, 605) mg with a range of 310–2813 mg to 331 (SD, 67) mg with a range of 207–538 mg (P < 0.0001 by a one-way, repeated-measures analysis of variance). Estimated dietary calcium intake remained stable throughout the study by the Bonferroni and Dunn test. As measured by the dietary questionnaire, the goal of reducing the daily calcium intake to less than 500 mg was met in 55 of 57 patient weeks.
Linear regression analysis demonstrated a trend toward higher urinary calcium excretion in patients with higher reported dietary calcium intakes (P = 0.06; R2 = 0.19) that was not statistically significant. The extent to which the reduced dietary calcium was important to minimizing the development of hypercalcemia in these patients will require further study and more objective measures of calcium intake.
The Phase I study was not designed to demonstrate efficacy, and 7 of the 15 patients did not have measurable disease. No patient had a documented response.
Five of eight patients with measurable disease had stable disease. Among them, an adenocarcinoma of the lung patient, an adenocarcinoma of the pancreas patient, and a hepatocellular carcinoma patient received 2 cycles of therapy and remained stable for the entire 16 weeks of their time on study. The hepatocellular carcinoma patient had an associated 70% decline in her serum α-fetoprotein level. The remaining three patients with measurable disease had evidence of progressive disease.
Four of seven patients without measurable disease had no evidence of progression. Among them, a breast carcinoma patient had a decline of the CA 15-3 from 445 U/ml to 365 U/ml at the end of the 8-week study period. It did not return to prestudy levels until after Week 22. Another prostate carcinoma patient received 2 cycles of therapy and had a stable PSA for the entire 16 weeks on study after experiencing a rapidly rising PSA before his enrollment. Whether these changes in serum tumor markers are indicative of an antitumor or biologic effect of calcitriol on these tumors cannot be determined without further studies. The remaining three patients without measurable disease had either tumor markers or clinical evidence of progressive disease.
One of the many normal functions of calcitriol is hormonal regulation of the blood calcium level. Physiologically, the production of calcitriol by the kidney from its precursor 25-hydroxycholecalciferol is tightly regulated. It is noteworthy that attempts to escalate the daily dose of calcitriol to take advantage of its known antiproliferative effect against cancer cells have been stymied by the development of hypercalcemia. Studies of the effects of calcitriol on tumor cell growth in vitro suggest to us that rather brief exposure of tumor cells to supraphysiologic concentrations of calcitriol can have prolonged antiproliferative effects. Further, a calcitriol-mediated increase in gastrointestinal absorption of calcium may precede calcitriol's effects on release of calcium from bone. We, therefore, postulated that an intermittent schedule of calcitriol combined with a reduced calcium diet should allow a substantial dose increase and, thereby, should improve the therapeutic index of this agent as an anticancer drug.
The results of this study do demonstrate that weekly pulse dosing of calcitriol allows substantial escalation of the total weekly dose without significant clinical toxicity. Doses up to 2.8 μg/kg were administered with minimal toxicity. However, no further increases in either peak plasma calcitriol levels or AUC were observed beyond the 0.48 μg/kg dose level. These pharmacokinetic results indicated that further escalation beyond 0.48 μg/kg did not result in increased exposure of malignancies to calcitriol and, therefore, was unlikely to result in any therapeutic benefit. The study was, therefore, terminated before any dose-limiting toxicity was observed.
Despite termination of the study before dose-limiting toxicity was reached, we were able to confirm the hypotheses of the study. Weekly pulse administration of calcitriol did result in a substantial increase in exposure to calcitriol without substantial toxicity. At doses of 0.48 μg/kg and beyond, the observed peak plasma calcitriol levels averaged 35-fold over baseline.
The dose of 0.5 μg/kg has been chosen for further Phase II clinical trials of calcitriol. At this dose, exposure to plasma calcitriol is substantially elevated as compared with baseline for both the 48-hour period after ingestion and for the entire week by an average of 14-fold and 4.8-fold, respectively. Thus, both peak and average exposure to calcitriol can be substantially increased by pulse dosing without causing clinical toxicity, including significant hypercalcemia.
The pharmacokinetic results including peak levels, AUC, and initial half-time disappearance for plasma calcitriol after ingestion of doses of 0.06 μg/kg are essentially identical to those reported by Papapoulos, et al.38 when they administered an oral dose of 4 μg calcitriol (0.057 μg/kg for a 70 kg subject) to normal volunteers. These parameters increased linearly with doses up to a dose of 0.48 μg/kg. Beyond this range, neither the peak calcitriol levels nor the AUC rose, whereas the initial half-time of disappearance at higher dose (1.6 μg/kg) appeared to be similar to that observed at lower doses.
At this time, the reason for the plateau in peak calcitriol levels and AUCs for doses beyond 0.5 μg/kg is unknown. Synthesis, absorption, and metabolism of calcitriol are each highly complex and normally tightly regulated to control calcium metabolism. In addition, the formulation of calcitriol in the current formulation contains large amounts of coconut oil and gelatin, thus further complicating the absorption. At this time, we cannot formally distinguish between the possibilities of either saturable absorption of calcitriol at higher dose levels versus an inducible and more rapid metabolism of absorbed calcitriol after administration of higher doses. However, we believe that the results appear to be more compatible with a saturable absorption process. In this regard, it should be noted that the calcitriol formulation used in this study, Rocaltrol, consists of 0.5 μg of calcitriol dissolved in coconut oil and contained in a gelatin capsule. Thus, doses of 0.5 μg/kg require ingestion of 1 capsule per kilogram. It is, perhaps, not surprising that such a large number of capsules and volume of coconut oil might have interfered with absorption at higher doses. Better bioavailability of high-dose calcitriol may be achievable with a formulation containing a larger and more concentrated dosage. Better bioavailability also may be achievable with alternative schedules of oral administration (i.e., a longer interdose interval on each day of therapy) or alternative routes of administration such as intravenous or subcutaneous.
This was a Phase I trial of escalating doses of calcitriol rather than a test of efficacy. No patient had a clinical response in bidimensionally measurable disease that fulfilled standard response criteria. Nonetheless, several patients had prolonged stable disease that might have indicated a biologic effect of calcitriol on the malignancy. Calcitriol can halt proliferation either through inducing cell death or through increased differentiation. Design of any future efficacy trials of calcitriol should take into account the possibility that it may produce a beneficial effect through disease stabilization rather than reduction of tumor size.
In summary, this short-term study demonstrated the feasibility of substantial escalation of the total weekly dose of calcitriol by adopting a pulse weekly administration schedule. Peak blood levels of calcitriol, approximately 25-fold above the upper limit of the physiologic range, were achieved with minimal toxicity. These levels were well within the range where antiproliferative effects of calcitriol are observed. Based on the observation that blood levels and AUC of calcitriol did not increase linearly with increases in doses beyond the 0.48 μg/kg level, we recommend a dose level of 0.5 μg/kg as an appropriate dose of calcitriol in Phase II trials of pulse weekly dosing. In addition to efficacy, initial Phase II studies will need to examine the toxicity of long-term administration of pulse calcitriol. Renal toxicity, bone loss, or other unexpected toxicity not seen in this trial may be encountered with long-term therapy. Subsequent studies also should examine the contribution of the reduced calcium diet to the control of hypercalcemia.
The authors thank Gary Sexton for his assistance with the statistical analysis.