Early Control of PTH and FGF23 in Normophosphatemic CKD Patients: A New Target in CKD-MBD Therapy?

Discussion

Previous studies have shown that increased serum PTH concentrations occur in patients with CKD stage 3 who typically have normal calcium and phosphate levels (3,7). The Study to Evaluate Early Kidney Disease (SEEK) (12), a cross-sectional analysis of 1814 patients with CKD stage 3 to 5, demonstrated that calcium and phosphate values did not become abnormal until GFR fell below 40 ml/min. However, nearly 60% of patients with GFR <60 ml/min had elevated PTH levels. Despite these findings, the current guidelines only recommend a dietary phosphate restriction for these patients. In this 6-wk, dose-titration study, we demonstrate for the first time that the use of phosphate binders in early CKD patients with normal serum phosphate is effective at lowering total urinary phosphate excretion, fractional phosphate excretion, and serum PTH without significant changes in serum calcium or phosphate levels. Thus, these data demonstrate that early phosphate binder use provides additional PTH control beyond that of a phosphate-restricted diet.

In study presented here, sevelamer hydrochloride treatment appeared more effective at lowering serum FGF23 levels. These findings cannot be attributed to differential control of serum phosphorus or 1,25-vitamin D₃, two known regulators of FGF23. It is possible that sevelamer's increased ability to reduce FGF23 occurs through an unknown effect. In addition, these findings also raise the possibility that increased calcium load may contribute to FGF23 elevation. Serum FGF23 levels have been found to correlate with serum calcium independent of serum phosphate in individuals with primary hyperparathyroidism who underwent parathyroidectomies (13). In addition, in vitamin-D-receptor null mice, dietary calcium supplementation significantly increased FGF23 mRNA abundance (14), indicating that calcium could be another determinant of FGF23 production. On the basis of these studies, further analyses of factors controlling FGF23 production are warranted.

The current understanding of the role of FGF23 is still emerging. Similar to the “trade-off hypothesis” (15) proposed for PTH, small elevations of FGF23 may have a beneficial effect by preventing hyperphosphatemia in early CKD. Over the long term, high FGF23 levels can favor the development of hyperparathyroidism because FGF23 inhibits the activity of 1α-hydroxylase, the rate-limiting enzyme responsible for 1,25-vitamin D synthesis (9). In a prospective study performed in dialysis patients with mild secondary hyperparathyroidism, serum FGF23 levels were significantly associated with the development of severe hyperparathyroidism (16). In an observational study, Fliser et al. (9) found FGF23 to be an independent predictor of disease progression in patients with nondiabetic CKD. Furthermore, FGF23 levels were a more accurate predictor of progression than serum phosphate, calcium, or PTH concentrations. Finally, a recent study has shown that FGF23 levels are independently associated with mortality among patients initiating hemodialysis treatment (10). The mortality risk associated with elevated FGF23 was substantially larger than that reported for serum phosphate. Together, these results suggest that the management of FGF23 may become an important goal in the treatment of CKD-MBD.

The significant increase in bone alkaline phosphatase and decrease in deoxypyridinoline that was found exclusively in the sevelamer group could be considered as an improvement in bone remodeling of these patients. However, bone biopsies were not performed, preventing definitive diagnosis of renal osteodystrophy and only a long-term study with bone biopsy analysis could confirm these findings. In a similar way, the significant decrease of 25-vitamin D should be seen cautiously, because no other study found a sevelamer-related effect on serum levels of this hormone.

To the best of our knowledge, this is the first clinical study that has evaluated the effects of phosphate binders on FGF23 in CKD stages 3 to 4. An experimental study involving uremic rats showed that the administration of sevelamer decreased FGF23 levels (17). Another study involving six dialysis patients demonstrated that the suspension of phosphate binders increased FGF23 levels, albeit not significantly (18). A Japanese study performed in 46 hemodialysis patients who were randomized to receive calcium carbonate alone or calcium carbonate and sevelamer hydrochloride showed that only the second group presented a significant decrease in serum FGF23 (19). The study presented here extends these data by demonstrating that a reduction in FGF23 can occur in patients without any changes in serum phosphate levels, suggesting that they do not adequately reflect phosphate balance. Similar observations have already been be made with serum calcium (20)

The current study presents several limitations. We excluded diabetic patients, and diabetic nephropathy is a main cause of CKD. However, in our study we preferred to exclude patients with a significant proteinuria, which would certainly be associated with vitamin D deficiency, and diabetic patients would certainly present a significant degree of proteinuria. A similar decision was taken by Fliser (9), who evaluated only young, nondiabetic CKD patients in the Mild to Moderate Kidney Disease (MMKD) study and found that FGF23 was a good predictor of CKD progression. Indeed, our group also recently found that FGF23 is a good predictor of CKD progression in diabetic patients (21). Nevertheless, in a long-term, prospective study diabetic patients will necessarily be included and the benefits of phosphate binders on serum PTH and FGF23 control must be confirmed in this population.

Our study is also limited by a small population size and the short duration. However, the current understanding of phosphate management in CKD leads us to propose a change in the paradigm: To move from the recommendation of the current guidelines [Kidney Disease Outcomes Quality Initiative (K/DOQI) and Kidney Disease: Improving Global Outcomes (KDIGO), which recognize the use of phosphate binders for CKD 2 to 4 patients only when serum phosphate is greater than 4.6 mg/dl] to a more aggressive control of phosphate load using phosphate binders even in the presence of normal serum phosphate values. A recent review published by Isakova (22) calls our attention to the need of a better understanding of the role played by phosphate and FGF23 in CKD, proposing to the nephrologic community a long-term study in this population. Certainly a long-term, randomized and probably multicentric study will be necessary to confirm this theory. In such a study, the long-term benefits of early binder treatment on vascular calcification, bone health, and disease progression would also be assessed. However, our preliminary findings showed that the use of phosphate binders in this population is not associated with serious adverse events in the short term. They also provided some preliminary data regarding PTH control and surprisingly showed that the FGF23 response could be distinct dependent on the phosphate binder that was used.

In conclusion, the use of phosphate binders in CKD patients with normal serum phosphate levels was associated with improved control of secondary hyperparathyroidism without corresponding changes in serum phosphate levels. Sevelamer was also able to decrease serum FGF23. The benefits of these findings should be confirmed in long-term studies.