The Effect of Cinacalcet on Bone Remodeling and Renal Function in Transplant Patients With Persistent Hyperparathyroidism

Abstract

Background. 

Parathyroidectomy is associated with renal functional losses in transplant patients; cinacalcet offers an attractive alternative.

Methods. 

We performed a prospective observational study in 58 patients with persisting hyperparathyroidism after renal transplantation (Ca²⁺≥2.6 mmol/L) and impaired renal transplant function (estimated glomerular filtration rate [eGFR] <50 mL/min). The patients received 30 to 90 mg cinacalcet for 12 months with the target to normalize serum Ca²⁺. We measured parathyroid hormone (PTH), serum Ca²⁺, serum phosphorus, alkaline phosphatase, bone-specific alkaline phosphatase, osteocalcin, and telopeptide at 0, 1, 2, 3, 6, 9, and 12 months of cinacalcet treatment. Fractional excretion of calcium and phosphorus (n=24) were monitored at 0 and 1 month.

Results. 

At inclusion, creatinine was 181±70 μmol/L, eGFR 43±19 mL/min, PTH 371±279 pg/mL, and Ca²⁺ 2.73±0.22 mmol/L. We observed nephrocalcinosis in 58% of biopsied patients at enrollment. After cinacalcet, Ca²⁺ decreased significantly and normalized at nearly any measurement. Phosphorus increased significantly at months 1, 9, and 12. PTH decreased significantly, but only at months 9 and 12 and did not normalize. Bone-specific alkaline phosphatase increased significantly (>normal) by month 12. eGFR decreased and serum creatinine increased at all time points. The Δ(creatinine) % increase correlated significantly with the Δ(PTH) % decrease at month 1 and 12. Telopeptide and alkaline phosphatase correlated with PTH and telopeptide also correlated with serum creatinine.

Conclusion. 

Calcium-phosphorus homeostasis in hypercalcemic renal transplant patients normalizes under cinacalcet and PTH decreases, albeit not to normal. The renal functional decline could be PTH mediated, analogous to the effects observed after parathyroidectomy.

In 20% of patients undergoing renal transplantation, parathyroid hormone (PTH) and serum calcium (Ca²⁺) levels remain increased at 1 year after surgery (^1–5). Others and we have observed that parathyroidectomy after renal transplantation bears the risk of short-term renal transplant deterioration in a considerable number of patients (^6–11). We noted that the renal transplant functional decline was correlated with the decrease in PTH after parathyroidectomy. We reasoned that PTH may have a positive influence on renal hemodynamics to uphold renal function (¹¹). Cinacalcet up-regulates the sensitivity of the parathyroid calcium-sensing receptor to calcium (^{12, 13}), lowers PTH, and improves calcium-phosphorus homeostasis in dialysis patients with uncontrolled secondary hyperparathyroidism (^14–17). Furthermore, cinacalcet even reduced bone turnover and tissue fibrosis in bone histology in many dialysis patients (¹⁸). However, cinacalcet is not yet established in the treatment of persisting hyperparathyroidism after renal transplantation. We conducted a prospective observational study of cinacalcet in patients with persistent hyperparathyroidism and hypercalcemia after renal transplantation. The patients whom we studied had some degree of impaired renal function, which made the avoidance of parathyroidectomy in these patients particularly desirable.

RESULTS

Fifty-eight renal transplant patients (33 men and 25 women) in our outpatient transplant unit met the criteria of low to medium renal transplant function (serum creatinine 181±70 μmol/L and estimated glomerular filtration rate [eGFR] 43±19 mL/min), persistent hyperparathyroidism (PTH 371±279 pg/mL), high Ca²⁺ (2.73±0.22 mmol/L), and nephrocalcinosis in the renal transplant biopsy (n=25 of 43 biopsied patients, 58%). Numerous patients had several criteria. One patient had pulmonary calciphylaxis. Parathyroidectomy had been performed in 8 of 58 patients before transplantation (14%; in 3 cases twice), who nonetheless met our criteria (listed earlier) and did not differ in serum calcium and PTH levels from the others. Parathyroidectomy in these patients had not been effective (³), or hyperparathyroidism had been recurrent after subtotal (³) or total (²) parathyroidectomy, the latter with autotransplantation.

The mean patient age was 55.1±9.1 years. The patients had passed a period of 6.9±3.5 years of dialysis treatment and of 3.8±4.4 years after renal transplantation. Immunosuppression was cyclosporine based in 32 patients (54%), tacrolimus based in 19 patients (33%), performed by mycophenolate mofetil and steroids in 5 patients (9%), and sirolimus based in 2 patients (3%). During the time from June 2005 to February 2008, the patients began cinacalcet treatment 30 to 90 mg per day for 1 to 12 months (10.1±3.6 months including nine drop outs, see below) with the target of lowering serum calcium to the normal range (2.3–2.6 mmol/L). In 4 of 58 patients, more than 30 mg of cinacalcet per day was given (mean 35.2±15 mg). The side effects of cinacalcet were mild and mostly transient gastrointestinal disorders; they led to cessation of the drug in only one patient after 3 months. Another patient, who had been transplanted 5 years ago and had well-known chronic tubulointerstitial changes in a preceding biopsy, stopped cinacalcet after 3 months because of renal function deterioration, which was suggested to be caused by the drug; renal function thereafter returned to the basal level. In a third patient, cinacalcet was stopped after 2 months during the course of an aortocoronary bypass operation with a prolonged postoperative course.

One patient had to be dialyzed because of transplant failure 3 months after start of the study (seven others within the year after the end of the study). Four patients died during the study (7%; one because of ileus after 1 month, one because of septicemia and one of unknown cause, both after 3 months, and one because of suspected myocardial infarction after 6 months). One patient had parathyroidectomy 10 months after start of the study because of persistent hypercalcemia (2.6–2.9 mmol/L) despite moderate PTH under 30 to 60 mg cinacalcet. Four others had parathyroidectomy within the year after the end of the study, one because of persistent hypercalcemia and three initiated by their nephrologist because of a high PTH.

Forty-three of 58 patients had at least 1 renal transplant biopsy during 3 years preceding inclusion to the study, including protocol biopsies and biopsies for cause (^{19, 20}). In 25 of 43 patients with renal biopsy (58%), nephrocalcinosis was seen. Only 3 of these 25 patients with nephrocalcinosis had hypophosphatemia, probably for they usually had renal insufficiency. We do not substitute phosphorus, because we know from former investigations that these patients get more nephrocalcinosis (²⁰). Patients with nephrocalcinosis nearly always were hypercalcemic (23 of 25 patients). The calcium deposits were located in the tubular lumina (¹¹) or in the tubulointerstitium (⁹) or both (⁵). In nine of these patients, the nephrocalcinosis was pronounced and in eight cases a repeated biopsy finding; in four cases, nephrocalcinosis was repeatedly severe. Only one of these four patients with repeated severe nephrocalcinosis had been parathyroidectomized before, which had not been effective. Patients with nephrocalcinosis had a lower eGFR than the others (39.2 vs. 44.5 mL/min); however, this was not statistically significant.

The results of measuring the aforementioned parameters are summarized in Table 1. Serum Ca²⁺ decreased after cinacalcet with high statistical significance at all time points. Serum Ca²⁺ normalized at any time point of measuring it in 53 of 58 patients (91%) and normalized in most but not at all time points in 4 of 58 patients (7%). Serum phosphorus increased regularly, but with statistical significance only at month 1 (P=0.03), 9 (P=0.02), and 12 (P=0.04). PTH decreased with treatment; however, statistical significance was not reached earlier than month 9 (P=0.03) and 12 (P=0.01). PTH was not lowered at all time points in 9 of 58 patients (16%) and reached a level of less than 150 pg/mL in only 34 of 58 patients (59%). Deterioration of renal function (Table 1) was observed regularly under treatment. Patients with cyclosporine-based immunosuppression had the same creatinine increase 1 month after cinacalcet intake (n=32, 162±67 before cinacalcet vs. 175±71 μmol/L at 1 month, P=0.02), as did the patients with other immunosuppression (n=27, 204±67 vs. 220±77 μmol/L, P=0.006). Alkaline phosphatase, osteocalcin, and telopeptide did not change significantly under cinacalcet treatment, whereas bone-specific alkaline phosphatase (BAP) increased significantly above normal but not earlier than month 12 (Fig. 2, P=0.03). Fractional excretion of calcium increased and that of phosphorus decreased under cinacalcet treatment, but without statistical significance (tested only once in the morning). Calcineurin inhibitor predose trough levels did not change significantly during the study (with the exception of a significantly lower cyclosporine trough level at month 9, Table 1).

TABLE 1: Results of measuring PTH, serum creatinine, eGFR, serum Ca²⁺ (adjusted for protein), serum PO₄, alkaline phosphatase, BAP, FE_Ca, and FE_PO4 without and with cinacalcet, Cy, and tacrolimus

FIGURE 2.: Intact parathyroid hormone (iPTH), alkaline phosphatase (AP), and bone-specific alkaline phosphatase (BAP) before and during the intake of cinacalcet (*P<0.05).

Alkaline phosphatase and telopeptide were correlated with PTH before cinacalcet treatment (r=0.5 and 0.6, respectively, P=0.001) and after treatment (month 12, r=0.3 and 0.6, respectively, P=0.04 and 0.005, respectively), whereas BAP correlated with PTH only before treatment (r=0.5, P=0.01). Telopeptide was correlated with serum creatinine before (r=0.5, P=0.01) but not after treatment. Bone- specific alkaline phosphatase and osteocalcin were not correlated with serum creatinine. Alkaline phosphatase, bone-specific alkaline phosphates, osteocalcin, and telopeptide were not correlated with serum Ca²⁺. The PTH concentration was not correlated with Ca²⁺, phosphorus, or serum creatinine. Δ(PTH)% decrease was correlated with Δ(creatinine)% increase in month 1 (r=0.4, P=0.01) and at month 12 (r=0.4, P=0.01); and with Δ(eGFR)% decrease at month 12 (r=0.3, P=0.04). Δ(Ca2+)% was not correlated with Δ(creatinine)% and Δ(eGFR)%.

DISCUSSION

The important findings in our study are that cinacalcet nearly invariably lowers serum Ca²⁺ to normal in hypercalcemic transplant patients with persistent hyperparathyroidism. PTH levels invariably decrease as well; however, this latter decrease requires months. Cinacalcet treatment also influences bone and matrix protein markers. Disconcerting is the fact that we observed a decreased eGFR with cinacalcet treatment, as others and we reported earlier in the patients undergoing parathyroidectomy (^6–11). These latter observations particularly are vexing; however, they require careful consideration. The Δ(PTH)% with cinacalcet was correlated with Δ(creatinine)% in our patients at month 1 and 12. These findings robustly suggest that these parameters could be related in a cause-and-effect fashion. Our study was (relatively) long term, and biopsy material was included, thanks to our protocol biopsy program.

Patients with nephrocalcinosis in biopsy nearly always were hypercalcemic. Because these patients in our series had an inferior graft function compared with the others, which corresponds to reports of the literature (²⁰), effective treatment of hypercalcemia and thereby possibly also of nephrocalcinosis seems extremely desirable. However, it is not clear, if really hypercalcemia is the only reason to induce nephrocalcinosis, or if chronic changes in the graft additionally allow calcium deposition in the tubular lumina and in the interstitial tissue.

BAP, which is the best marker of bone formation (^{21, 22}), was the only biomarker of bone remodeling that increased significantly above normal, albeit by month 12 after starting cinacalcet medication. Cinacalcet has recently been shown by bone histomorphometric analysis in 10 cases to decrease bone formation rate while increasing the osteoblast surface (“minimodeling” [^{23, 24}]). Before cinacalcet was begun, bone-specific alkaline phosphatase was correlated with PTH but no longer at month 12. Alkaline phosphatase and telopeptide were correlated with PTH before and after treatment. Moreover, telopeptide was correlated with renal function, which means that we cannot use it for measuring bone resorption in this setting. On the whole, our findings support the notion that our variables could indicate some kind of bone remodeling in these patients. Thus, we have a significant sign of increased bone remodeling under cinacalcet in our patients; however, we do not have a reassuring marker reflecting decreased bone resorption.

The immediate effect of cinacalcet on calcium phosphorus homeostasis has been reported by others (^25–38). Their and our data suggest that the rapid serum Ca²⁺ decrease under cinacalcet not only depends on PTH but also could depend on other more direct effects of cinacalcet on the renal tubular system or directly on the effect of cinacalcet on the calcium sensing receptor (³⁹). Other authors report more complex short-term effects of cinacalcet during 24 hr after intake. PTH levels decrease during 12 hr significantly; however, they increase to baseline levels after 12 hr (⁴⁰). The serum Ca²⁺ remains suppressed during 24 hr, which is accompanied by a significant increase of fractional urinary calcium excretion and a significant decrease of fractional urinary phosphorus excretion, especially during the first 8 hr (^{28, 39, 41}). Thus, these parameters measured only in the morning do not reflect these short-term changes. We do not know the effects of the short-term changes on bone remodeling. The increase of bone-specific alkaline phosphatase values suggests that there is a change in bone remodeling after 1 year of treatment with cinacalcet, even if the results of bone biopsy after cinacalcet treatment reported recently are difficult to interpret (²³).

There are contradictory reports on the reaction of renal function on cinacalcet treatment, while a possible deterioration of renal function after parathyroidectomy seems to be a well-established effect (^6–11). Most authors report a “stable” renal function under cinacalcet (^{27–33, 35–38, 41}). However, other authors reported a reduction of renal function under cinacalcet, as we reported in this study (^{25, 34, 42}). This decrease was generally reversible after cessation of the drug (²⁶). According to our findings, the situation after renal transplantation in patients under cinacalcet treatment in general resembles that after parathyroidectomy: namely, the PTH decrease is associated with renal function deterioration, and these parameters even correlate (¹¹). We found that the ΔPTH decrease correlated with the Δserum creatinine increase at month 1 and 12 and also with the ΔeGFR decrease at month 12. However, with cinacalcet, the lowering of PTH was much more moderate than after parathyroidectomy, which paralleled any renal functional decline. This observation supports experimental studies suggesting that PTH has a regulatory influence on renal perfusion, eGFR, and mesangial cell function (^43–46). Infusion of PTH in rats and humans leads to an increase of renal plasma flow and GFR, probably due to an hemodynamic effect on the renal vascular system (^{45, 46}). Falck et al. (⁴²) observed an interaction between cinacalcet and the secondary cyclosporine metabolite AM19; the increase in the latter correlated with decreasing GFR. In their report, only patients under cyclosporine showed a renal function decline, not patients with other immunosuppression. Our findings do not support this hypothesis, because in our study, patients with cyclosporine-based immunosuppression had the same degree of creatinine increase as the other patients.

Our study has numerous limitations. We reason that a randomized controlled trial with hard outcomes in posttransplant patients with cinacalcet would be desirable, but it is difficult to perform. Such a study is being performed in dialysis patients, whose numbers are far greater than those reflected in the transplant community. Only a limited number of transplant patients actually would need cinacalcet. However, we found that the drug seems to be safe and well tolerated. After renal transplantation, the drug was more successful in correcting serum Ca²⁺ (essentially all patients) and phosphorus levels, than in normalizing PTH, although the levels were positively influenced. We do not know which of these clinical parameters is the more important. It must be taken into consideration that we measured trough level PTH in the morning before taking cinacalcet, which seems to reflect the highest value before the lowering effect of cinacalcet (^{40, 41}). We remain concerned about possible PTH-related extraosseus (soft tissue) calcifications with their potentially disastrous cardiovascular components (^{15, 47–49}). A remaining issue is whether cinacalcet could preserve the positive effect of high PTH values on renal hemodynamics, at least in patients with impaired renal function. We have also not answered the issue of whether cinacalcet is able to prevent and to reverse parathyroid hyperplasia over the long term, which remains unproven (^{28, 50–53}). The effect of cinacalcet on nephrocalcinosis could not be measured, because we had only few biopsies during and after the study (protocol biopsies in our department are performed in the first 6 months after transplantation). Preventing nephrocalcinosis would have a positive influence on renal function.

The perspectives of our findings include the possibilities that cinacalcet could replace parathyroidectomy in many hypercalcemic patients with persistent hyperparathyroidism after transplantation. We were concerned that a decrease in eGFR that was correlated with drug response was observed in these patients. We cannot explain these findings mechanistically; however, they could be related to hemodynamic eGFR effects related to PTH. Obviously, the issue requires further observational studies, because a randomized controlled prospective trial may be difficult to perform because of the low number of cases. In the mean time, we suggest that cinacalcet is a desirable alternative in hypercalcemic transplant patients with continued hyperparathyroidism.

MATERIALS AND METHODS

The transplant patients in our unit are invited to participate in a protocol biopsy program (¹⁹). They were also queried about willingness to participate in any clinical research initiatives of our program. The Hannover Medical School internal review board approved the study, and written consent was obtained from all participants. Renal transplant patients with persisting hyperparathyroidism, high Ca²⁺ (≥2.6 mmol/L) and pronounced nephrocalcinosis on renal transplant biopsy (by protocol after 6, 12, and 26 weeks or by biopsy per indication), and impaired renal transplant function, eGFR after Cockcroft and Gault less than 50 mL/min, were treated with 30 to 90 mg cinacalcet for 1 to 12 months (10.1±3.6 months). The target was to lower Ca²⁺ to normal levels. Thus, the dosage of cinacalcet was adjusted accordingly. We measured serum creatinine and calculated eGFR (⁵⁴), PTH, Ca²⁺, serum phosphorus, alkaline phosphatase, and calcineurin inhibitor trough level at 0, 1, 2, 3, 6, 9, and 12 months after the start of cinacalcet (58 patients). We included BAP, osteocalcin, and carboxy-terminal telopeptide in serum (35 patients). We monitored fractional excretion of Ca²⁺ and phosphorus (FE_Ca and FE_PO4; Fig. 1) before and after 1 month after the start of cinacalcet medication (24 patients). Blood and urine were obtained in the morning before taking medication. Serum Ca²⁺ was adjusted for serum protein.

FIGURE 1.: (a) serum Ca²⁺, (b) serum phosphorus, (c) intact parathyroid hormone (iPTH), (d) serum creatinine, and (e) estimated glomerular filtration rate (eGFR), before and during cinacalcet. The bars represent the mean values (for standard deviation, see Table 1). The asterisks indicate that the comparison with month 0 (before cinacalcet) was significant (*P<0.05; **P<0.01; and ***P<0.001).

PTH was measured using an immunochemiluminometric assay (Liaison N-tact PTH-Assay, DiaSorin; normal range 10–65 pg/mL). BAP was measured by immunoenzymetric assay (normal range 7–20 μg/L; Fig. 2), osteocalcin by chemiluminescence immunometric assay (normal range <2.0–21 μg/L), and carboxy-terminal telopeptide by radioimmuno assay (normal range 2.1–5.6 μg/L in women and 2.1–5.0 μg/L in men). Values are given as means±standard deviation. Differences between mean values were tested for statistical significance by the unpaired Student's t test. Correlations were calculated by linear regression (Y=a+b×X). Differences between relative frequencies and percentages were compared by chi-square testing. The significance level was P less than 0.05.

ACKNOWLEDGMENT

The authors thank Mr. Horst Doehring for his essential help in managing the study.