Four decades of phosphate management
Four decades ago, Professor Slatopolsky discovered that serum phosphate levels in- creased as CKD progressed and as the GFR diminished. He also showed that parathyroid hormone increased as GFR declined, but this was improved when phosphate intake was reduced in dogs with CKD. These data are still relevant today, but we now know more of the mechanisms involved in hyperphosphatemia in CKD, including FGF-23, klotho, vitamin D, and parathyroid hormone.
There are various phosphate binder options for the treatment of hyperphosphatemia, the risks and benefits of each vary (Figure 5):
- Aluminum-containing phosphate binders are very efficacious, but the aluminum can be absorbed by the body, accumulate and cause toxic side effects, such as osteomalacia, encephalopathy and microcytic anemia.
- Calcium-containing binders are efficacious, but excessive calcium can be absorbed, leading to vascular calcification.
- The non-calcium metal-based binder lanthanum carbonate is absorbed and can accumulate in the body. Long-term studies are required to investigate lanthanum’s toxicity and its effect on vascular calcification.
- The metal-free, non-absorbed phosphate binder sevelamer has been shown to reduce the progression of vascular calcification and to lower mortality in CKD patients in some studies. Sevelamer also improves bone quality, reduces inflammation, and lowers lipid levels and oxidative stress.
Although dietary phosphate restriction in dogs with CKD reduced parathyroid hormone levels, reducing protein intake to reduce serum phosphate levels increases mortality, probably due to malnutrition.
The Treat-To-Goal and RIND studies showed that treatment with phosphate binders reduces progression of vascular calcification. According to the KDIGO guidelines, the presence of CV calcification strongly predicts CV morbidity and mortality in patients with CKD.
Can phosphate binders improve morbidity and mortality by slowing the progression of CV calcification? The data are not conclusive yet; more randomized clinical trials are necessary.
Questions raised included the ideal combination of existing phosphate binders, and whether there was a place for inhibitors of GI sodium/ phosphate co-transporters in the management of hyperphosphatemia; both nicotinamide and niacin reduce serum phosphate levels in hemodialysis patients. Randomized controlled trials are needed to investigate the effects of phosphate binders on hard outcomes in CKD stages 4–5.
Some phosphate binders, but not all, have been shown to reduce progression of vascular calcification.
Several new therapeutic options have recently been examined. The use of chitosan (a synthetic polymer that can be administered in chewing gum and can drastically lower phosphate levels in saliva), which results in a 2 mg decrease in serum phosphate, may be of therapeutic use in conjunction with other treatments.
Cinacalcet can only be given to patients with SHPT (only approximately one-third of patients with CKD), and it may impact phosphate, especially when used with a phosphate binder. A combined calcium/magnesium binder has recently been marketed in some countries, and although magnesium inhibits vascular calcification, it can also reduce bone mineralization. Several studies on iron-containing phosphate binders are ongoing, and results on iron release and accumulation are awaited with interest.
Phosphate and parathyroid hormone
It is well established that parathyroid hormone is the main component of the control system for modifying phosphate excretion in uremia, and that decreased phosphate intake can block the increase in parathyroid hormone in animal models of CKD. Phosphate can act via different mechanisms, such as reducing calcitriol production and release of calcium from bone; it also increases FGF-23 and klotho, which increase parathyroid hormone expression (Figure 6).
Animal studies in moderate CKD demonstrated that elevated parathyroid hormone levels fail to control serum phosphate, and that high phosphate levels reduce the calcium response to parathyroid hormone and the inhibition of parathyroid hormone by calcitriol.
In vitro studies have shown that the effect of phosphate on parathyroid hormone secretion is direct, concentration-dependent, requires cell contact (in whole parathyroid glands, slices, or collagen-matrix cultures, not dispersed cells), and is dependent on increased gene expression and enhanced stability of parathyroid hormone mRN A.
High phosphate concentrations interfere with the signalling pathway of the calcium-sensing receptors, preventing the release of arachidonic acid from the cell membrane. Results from several in vitro experiments have shown that the stimulatory effect of phosphate on parathyroid hormone secretion may be due to phosphate reducing the intracellular release of calcium in response to extracellular calcium, thereby reducing the inhibitory arachidonic acid signal.
Discussion on the fact that only intact cells (not dispersed cells) respond to phosphate led to an explanation that triggering parathyroid hormone secretion in one cell may cause it to communicate via signals from cell to cell so that more cells are triggered to produce more parathyroid hormone; phosphate may prevent this from occurring. In vitro studies have shown that 2-4 hours are required to see an effect, which indicates that phosphate may need to enter the cells first and act intracellularly. In vivo studies have shown that suppression of parathyroid hormone can occur within 15 minutes, suggesting rapid signalling from something other than phosphate.
Phosphate and FGF-23
The systems controlling phosphate homeostasis include bone, GI tract, parathyroid glands, and kidneys, and the hormones parathyroid hormone and FGF-23, which act on the kidneys to downregulate the sodium/phosphate co-transporter and thus increase phosphate excretion (Figure 7). FGF-23 increases as GFR decreases, and is much more sensitive to changes in GFR than phosphate or parathyroid hormone. Increased FGF-23 levels can predict progression of kidney disease in CKD and diabetic patients, with high levels of FGF-23 indicat- ing morbidity, such as left ventricular hypertrophy, and mortality.
In vivo studies conducted to investigate the effects following removal of FGF-23 showed increases in serum phosphate levels to be similar in CKD rats receiving anti-FGF-23 antibodies compared with normal rats.
Interestingly, the increase in urinary excretion of phosphate in CKD rats stopped following administration of anti-FGF-23 antibodies. Anti-FGF-23 antibodies increased 1,25(OH)2D production to the same level in CKD rats as in normal rats, while levels in CKD rats not receiving antibodies remained low.
These effects were mediated by a correction in the activity of the 1α hydroxylase enzyme, and a reduction in the activity of 24 hydroxylase enzyme. These in vivo results showed that FGF-23 is responsible for renal phosphate reabsorption and normalization of blood phosphate levels in mild CKD, and that FGF-23 acts independently of parathyroid hormone and reduces 1,25(OH)2D production, which leads to low calcium and increased parathyroid hormone; findings that are consistent with those in humans.
Measurement of biologically active FGF-23 in the circulation can be performed equally well using intact FGF-23 or the C-terminal portion of FGF-23. With reference to the trade-off hypothesis, the actions of phosphate on parathyroid hormone are probably dependent on elevated FGF-23 being secreted from the bone, resulting in lower 1,25(OH)2D levels, consequently triggering the release of parathyroid hormone.
Investigation into high levels of FGF-23 in the circulation and the off-target effects of this molecule are warranted. Interventions need to reduce the absorption of phosphate from the GI tract, and focus on patients with earlier stages of CKD to promote phosphate excretion into urine.
FGF-23 is responsible for renal phosphate excretion and maintenance of blood phosphate levels in mild CKD The presentation triggered questions as to the effect of increasing intake of dietary phosphate on FGF-23. When phosphate intake is increased in humans, plasma FGF-23 increases by only a small amount, whereas urinary phosphate excretion increases by a large amount. Interestingly, serum phosphate increases following parathyroidectomy in rats with CKD, despite the powerful effects of FGF-23; increasing FGF-23 does not control serum phosphate in SHPT or PHPT.
One suggestion was that 1,25(OH)2D is low in the absence of parathyroid hormone and is critical for the synthesis of FGF-23. In an organism that has little 1,25(OH)2D and is hypocalcemic, further reducing calcium absorption via FGF-23- mediated inhibition of 1,25(OH)2D would be harmful. In PHPT, FGF-23 is not as high as it should be as it is trying to conserve 1,25(OH)2D and calcium.
To date, klotho has not been measured and FGF-23 levels have not been detected in healthy persons with normal kidney function. The off-target effects of FGF-23 in klotho deficient mice include interaction with a variety of receptors or splice variants, although these studies are difficult because these mice also have severe hyperphosphataemia. No difference could be seen in klotho+FGF-23 knockout mice compared with klotho only knockout mice; therefore, an off-target effect of FGF-23 may not be detectable in mice.
Regarding the physiological function of FGF-23, some data suggest that, if the signal from the osteocyte tells the kidney to lose phosphate, the signal to the osteocyte comes from the dying kidney, but further studies are needed. A recent study shows that, in very early CKD, patients who have completely normal GFR have very high FGF-23 levels, totally normal parathyroid hormone levels and subtle reductions in serum phosphate; therefore, CKD itself may increase FGF-23. It has also been shown that, following kidney transplant into a CKD patient, FGF-23 levels decrease by 85% within a day.