Does Vitamin D Receptor and Calcium Receptor Activation Therapy Play a Role in the Histopathologic Alterations of Parathyroid Glands in Refractory Uremic Hyperparathyroidism?

Materials and Methods

Results

Table 1 shows the demographic, clinical, and biochemical characteristics of the entire group of 22 patients. Table 2 shows the dosage and the duration of treatment with cinacalcet and intravenous calcitriol (group A) and intravenous calcitriol only (group B).

No statistically significant difference was found when comparing the demographic, clinical, and biochemical characteristics of the three groups of patients (Table 3). The total number of pathologic glands removed was 82: 23 were ablated in the six patients of group A, 22 in the six patients of group B, and 37 in the 10 patients of group C (Table 3). Sixty-eight glands showed NH (82.9%). NH was more frequent in groups A (22 [95.6%] of 23 parathyroids) and B (21 [95.5%] of 22 parathyroids) than in group C (25 [67.6%] of 37 parathyroids; P < 0.05; Table 3).

The estimated weight of the parathyroids (expressed as the mean weight of individual parathyroid glands from each patient) was not significantly different when comparing the glands of the three groups (Table 3). Furthermore, the mean weight of the 68 glands showing NH was 1.45 ± 0.3 g, whereas that of the 14 glands showing DH was 1.26 ± 0.4 g. The difference was not statistically significant.

Oxyphil/chief cell ratio was significantly different when comparing the glands of the three groups (0.65 ± 0.3 in group A, 0.36 ± 0.4 in group B, 0.19 ± 0.1 in group C; P < 0.005; Table 3). Furthermore, Bonferroni post hoc test demonstrated that a significant difference existed specifically between groups A and C (P < 0.004; Table 3).

Histopathology of parathyroid glands (NH versus DH) was significantly predicted by BMI (patients with BMI >25 kg/m² had a probability 5.53 times higher than patients with BMI ≤25 of getting NH; P < 0.047), by the treatment groups (patients of group A had a probability 0.29 times higher than patients of group C and 0.14 times higher than patients of group B of getting NH; P < 0.007), and by gender (female patients had a probability 0.079 times higher than male patients of group C of getting NH; P < 0.031; Table 4).

The best fitted stepwise multiple regression analysis model (R² = 0.515; P = 0.0001) demonstrated that cinacalcet treatment (B = 0.362; SE = 2.586; P = 0.023; 95% confidence interval 0.060 to 0.665) was the only predictor of the oxyphil/chief cell ratio. All of the other variables tested in the regression model resulted in no association with the ratio (Table 5).

Discussion

As a premise to the discussion, we stress that we had the virtually unique opportunity of examining parathyroid glands of uremic patients who had never been exposed to calcitriol and/or cinacalcet in the attempt to treat SHPT. This opportunity was given by the peculiar type of recruitment of these patients for PTx, as already anticipated in the Materials and Methods section. Thus, patients of group C look like an experimental model of natural history of SHPT not exposed at all to any receptor therapy.

Our data show the following:

1. a statistically significant difference in the prevalence of NH when comparing patients who were treated with P binders only with patients who were treated with the receptor therapy (intravenous calcitriol and cinacalcet, either as monotherapy or in combination); furthermore, a stepwise forward logistic regression model showed that NH was significantly predicted by the treatment groups, the probability of NH being higher in patients who were treated with the receptor therapy (intravenous calcitriol and cinacalcet, either as monotherapy or in combination)

2. a statistically significant increase of oxyphil/chief cell ratio from 0.19 of the patients who were treated with P binders only to 0.65 of the patients who were treated with the combined therapy intravenous calcitriol and cinacalcet; however, the predominant role in this case is played by cinacalcet, as shown by the stepwise multiple regression analysis model

3. a statistically significant increase of probability of NH in female gender and in patients with a higher BMI

As far as item 1 is concerned, our data in a new set of patients confirm our previous findings that calcitriol pulse therapy was associated with an increase in the prevalence of NH of parathyroid glands (11). Our data show that also cinacalcet therapy is associated with a very high prevalence of NH when combined with intravenous calcitriol; however, when trying to determine the possible contribution of cinacalcet and/or calcitriol in the high prevalence of NH, a logical reasoning leads us to argue that the main role should be played only by intravenous calcitriol for the following reasons: (1) The latter is associated with a prevalence of NH of 95.5% (group B), whereas the addition of cinacalcet to calcitriol treatment does not seem to modify this prevalence (95.6%, group A); (2) calcitriol treatment preceded by many months the prescription of cinacalcet in all patients of group A; consequently, it is logical to think that the high prevalence of NH observed in group B (calcitriol treatment) would be observed also in group A at the moment of adding cinacalcet to calcitriol. It is obvious to say that logical evidence does not constitute scientific evidence. Thus, it is possible that also cinacalcet treatment per se might be associated with a high prevalence of NH. The only way to demonstrate the latter issue is to study patients who are treated with cinacalcet only. As well known, calcitriol has an antiproliferative action on many target tissues, including parathyroid glands; however, a secondary adverse effect at the pharmacologic doses used could be produced because of the antiproliferative action of calcitriol: The development of cell clones having functional autonomy in hyperplastic parathyroid gland. In that way, the prevalence of NH increases from 67.6% of group C to 95.5% of group B. The reasons for this paradoxic adverse effect could be looked for in the frequently occurring delay in starting calcitriol treatment, in a discontinuous instead of a continuous strategy of treatment, and in the use of pharmacologic instead of physiologic doses, as it should occur in every endocrinologic substitution therapy; however, two other possibilities should be considered for explaining the different prevalence of NH among the three groups: (a) A selection process could have occurred in groups A and B, which received calcitriol, in which calcitriol-sensitive parathyroid glands responded to treatment and less responsive glands affected by NH did not respond, resulting in a preponderance of glands with NH at the time of PTx. In this scenario, calcitriol treatment could not induce NH but could select for the presence of NH because of the decreased responsiveness to calcitriol associated with NH; and (b) the hypothesis cannot be excluded, despite that the three groups were similar at the time of surgery as far as serum Ca, P, and iPTH were similar, that group A patients were affected by the most severe form of SHPT (association of three drugs) when compared with group C patients (one-drug regimen). In other words, patients who were treated unsuccessfully with calcitriol could be those who presented the most aggressive histopathologic patterns of parathyroid hyperplasia already before starting calcitriol treatment.

The novel finding shown by item 2 is that cinacalcet is associated with a statistically significant increase in the oxyphil/chief cell ratio. Also in this case, when trying to determine the possible contribution of cinacalcet and/or calcitriol to the high oxyphil/chief cell ratio shown by group A patients, a logical reasoning leads us to argue that the main role should be played only by cinacalcet for the following reason: Calcitriol treatment preceded by many months the prescription of cinacalcet in all patients of group A; consequently, it is logical to think that the ratio of 0.36 observed in group B patients (calcitriol treatment) would be observed also in group A patients at the moment of addition of cinacalcet to calcitriol; this ratio went up to 0.65 after the introduction of cinacalcet therapy. Furthermore, a stepwise multiple regression analysis showed unequivocally that the predominant role was played by cinacalcet. We have to keep in mind three background points: (a) We and others have shown that oxyphil cells are well represented in NH (9,11). Recently, it was shown that oxyphil cells produce actively PTH-related protein and that oxyphil cells have a level of proliferation lower than that of chief cells (14); PTH-related protein could act through an autocrine/paracrine mechanism to regulate proliferation and differentiation of parathyroid cells (14), favoring their metaplastic change into the oxyphil phenotype (14,15); (b) CaR are expressed in many tissues, with the highest density being found in the chief cells of the parathyroid glands (2); (c) furthermore, very recently, it was also shown that CaR activation by the calcimimetic R-568 accelerates chief cell death, probably through an apoptotic mechanism in uremic rats in vitro (16). Thus, the hypothesis could be made that cinacalcet counteracts SHPT mainly through two synergistically operating mechanisms: An antiproliferative and a proapoptotic action on chief cells. In this way, the oxyphil/chief cell ratio increases, probably because of a more significant decrease in the absolute number of chief cells than because of an increase in the absolute number of oxyphil cells. The paradox that a seemingly beneficial effect, such as the increase in the oxyphil/chief cell ratio in the cinacalcet treatment group, was not able to preserve these patients from PTx could be explained by the short cinacalcet treatment time (3.3 ± 1.8 mo; range 2 to 7 mo) with a probably inadequate dosage titration in some cases and/or by the fact that the start of cinacalcet treatment was too delayed.

As far as item 3 is concerned, our data expand previous findings of other authors that female gender favors the development of refractory SHPT and the need for PTx (17–19). They also confirm our previous findings that female gender (10,20) and BMI were associated with an increased risk for NH of parathyroid glands (10). We demonstrated in that prospective study that a higher BMI was associated with a larger body burden of P in female hemodialysis patients, influencing by that way the severity of SHPT (10). Why female hemodialysis patients with high BMI may have a larger P body burden is a matter of speculation that we largely discussed in that article (10).

Conclusions

The development of a refractory SHPT continues to represent an important problem because of the frequent need for PTx. We found an association between calcitriol and/or cinacalcet therapy and a high prevalence of nodular type parathyroid hyperplasia and between cinacalcet therapy and high oxyphil/chief cell ratio. The meaning of the observed association between treatment modality and histopathologic changes of parathyroid tissue in refractory SHPT remains uncertain. Further prospective studies are required to confirm these findings and understand the underlying mechanisms.

Disclosures

None.