Effect of Increasing Doses of Cystine Binding Thiol Drugs on Cystine Capacity in Patients with Cystinuria

Abstract

Appropriate dosing of cystine binding thiol drugs (CBTDs) in the management of cystinuria has been based on clinical stone activity. When new stones form, the dose is increased. Currently there is no method of measuring urinary drug levels to guide titration of therapy. Increasing cystine capacity, a measure of cystine solubility, has been promoted as a method of judging the effects of therapy. In this study, we gave increasing doses of tiopronin or d-penicillamine, depending on the patients’ own prescriptions, to ten patients with cystinuria and measured cystine excretion and cystine capacity. The doses were 0, 1, 2, 3 grams per day, given in two divided doses, and administered in a random order. Going from 0 to 1g/day led to an increase in cystine capacity from −39.1 to 130.4 mg/L (P<0.009) and decreased 24 hour cystine excretion from 1003.9 to 834.8 mg/day (P=0.039). Increasing the doses from 1 to 2 to 3 g/day had no consistent or significant effect to further increase cystine capacity or decrease cystine excretion. Whether doses higher than 1 g/day have additional clinical benefit is not clear from this study. Limiting doses might be associated with fewer adverse effects without sacrificing a benefit of higher doses, if higher doses do not offer clinical importance. However, trials with stone activity as an outcome would be desirable.

Introduction

Cystinuria is a rare, autosomal recessive, inherited disease that leads to the development of recurrent kidney stones.[1] Affected individuals have onset of stones that often begins in teenage years, with relatively few options for treatment. Increased fluid intake, restriction of dietary animal protein and sodium intake, and urinary alkalinization constitute the first line of preventive therapies. When cystine stones recur despite these measures, we prescribe cystine binding thiol drugs (CBTDs) which increase the solubility of cystine in the urine. Currently available CBTDs are tiopronin (alpha-mercaptopropionylglycine) and D-penicillamine. These medications have been demonstrated to be efficacious in a limited number of non-randomized, non-controlled clinical trials.[2,3]

The studies in which CBTDs have been utilized have not systematically explored the clinical responses of patients to varying dosages. A range of doses is provided in these agents’ package inserts, but a method of determining an optimal dose has not been available. We have previously suggested that a preferred method of judging the effect of CBTDs is measurement of cystine capacity, currently a proprietary test developed at the University of Chicago and Litholink (Chicago, IL).[4,5] The test, in essence, is a test of cystine solubility, rather than a measure of cystine excretion. Questions have been raised about the ability of both colorimetric and liquid chromatography/mass spectroscopy (LC/MS) assays to accurately measure cystine excretion in the presence of thiol agents.[4] Cystine capacity overcomes these doubts and offers a different therapeutic strategy, by providing a measure of what is ultimately the goal of successful therapy: an increase in the capacity of urine to hold cystine in a soluble state. Although not yet definitively demonstrated, this value perhaps holds more clinical significance than 24 hour cystine excretion.

We previously demonstrated that CBTDs do increase cystine capacity, and lower the urinary supersaturation of cystine.[6] We would expect, and in other pending studies, we strive to show, that this effect will be associated with clinical benefit. In that previous study, we showed that cystine capacity increased at whatever dose of CBTD the participants were prescribed, compared with no CBTD. We now sought to determine the most effective dose of CBTDs by measuring cystine capacity in individuals taking progressively higher doses of the medications. Our hypothesis was that higher doses of CBTDS would yield more positive cystine capacity values.

Methods

We performed a crossover trial in which patients received escalating dosages of CBTDs. We enrolled 15 participants; 10 completed the study. Patients with cystinuria, ages 18–80 years, were recruited from the clinical practice of the principle investigator at routine clinic visits at Lenox Hill Hospital and, later, New York University Medical Center. The study was approved by the investigational review boards of both centers, and performed under the aegis of the Rare Kidney Stone Consortium, a member of the Rare Disease Clinical Research Center. The study was registered at Clinicaltrials.gov (NCT).

Eligible patients were taking CBTDs, either tiopronin or d-penicillamine, previously prescribed for the appropriate clinical indication. Therefore, in order to ensure safety, no patient would receive a prescription for either drug having not been previously exposed to it. All patients were judged to be tolerating their current therapy. After signing informed consent, the patients had a screening interview to determine that they did not have symptoms of renal colic or pending scheduled urologic procedures, either of which, if present, would have excluded their eligibility. The participants’ most recent bloodwork (including a complete blood count and comprehensive metabolic profile) and most recent urinalysis were reviewed to demonstrate that they had not experienced adverse effects of the medications. The remainder of the study protocol, including changes in doses and collection of urine, was performed by participants at home.

Each participant completed four periods of the study. The order in which each participant performed the four parts was assigned by computer, generating a random order of 1–4, in order to prevent a sequence effect. The four periods included 7 days at each of four doses of the CBTD they were taking before enrollment. On day 7 of each of the periods, a 24 hour urine collection was performed. The four phases consisted of the participants taking 0g of medication, 1g per day, 2g per day, and 3g per day. Each dose was divided into twice-a-day administration. There was no specified wash-out period in between study periods due to the relatively short half-life of the CBTDs. Some participants did the four study periods in consecutive weeks but that was not a requirement. With the exception of the study CBTDs, the patients continued all of their regular medications including alkali therapy. No changes in prescriptions of potassium citrate were made and doses of alkali were not standardized.

Patients continued their self-selected diets. Participants kept a detailed food diary for 48 hours: during the 6^th day of the first period (the day before the first 24 hour urine collection) and on the 7^th day of the first period, the day of the first 24 hour urine collection. We then asked them to replicate the diet consumed during those two days, during each of the subsequent periods of the study, repeating the same diet on the day before and the day of the 24 hour urine collection. We collected the food diaries at the end of the study to ensure compliance and to keep a record of what the subjects were eating. However we did not analyze the data from the food diaries; their purpose was to enable replication of the diet during the urine collections.

Participants had repeat blood testing (for complete blood count and comprehensive metabolic profile) and urinalysis within one month of completing the study to monitor for any potential adverse effects of changing the dosages of medication. Subjects did not receive compensation for participation in the study.

Urine collection procedures and measurement

An antibacterial preservative and a volume marker were added to the collection container at the beginning of the collection.[5] Patients removed a 50ml aliquot and alkalinized the rest of the collection with Na₂CO₃ to bring any precipitated cystine into solution.[7] They removed a 50ml aliquot from the alkalinized collection, and sent the aliquots of the original urine and the alkalinized sample to the Litholink laboratory by overnight delivery. Urine samples were also analyzed by standard analytic techniques for concentration of sodium, potassium, citrate, chloride, urea nitrogen, creatinine and pH with a Beckman-Synchron CX5 (Beckman Instruments, Brea, California).[8]

Cystine assay

Cystine was measured as previously reported.[9,7,4] A 100 ml portion of urine was diluted with 400 ml H₂O and 1 ml phosphate buffered saline buffer, pH 7.4 (Sigma-Aldrich Co., St Louis, MO, USA). A 300 ml portion of a 10% sodium cyanide solution was added and the mixture incubated at room temperature for 20 min. Addition of 100 ml of a 20% sodium nitroprusside solution initiates a colorimetric reaction linearly related to the concentration of cystine present. Absorbance was measured within 20 s of the nitroprusside addition at 521nm using a Beckman DU 650 spectrophotometer and concentration calculated from a standard curve run with each assay. The intra-assay coefficient of variation is 2.9% and the inter-assay coefficient of variation 2.6% in this laboratory.

Cystine capacity assay

The cystine capacity of each urine sample was measured by the solid phase cystine assay.[5] To 25 ml of urine, maintained at the patient’s pH, a known amount of solid-phase cystine (Sigma-Aldrich Co.) was added and incubated for 48 h at 37°C with constant stirring. The solid phase was then harvested by centrifugation at 3800 r.p.m. for 20 min at room temperature. The supernatant was removed and the remaining pellet dissolved in 25 ml of high-pH buffer (0.1M sodium carbonate, pH 9.9). Cystine concentration was determined in the supernatant and high-pH buffer as described above. The sum of cystine in the supernatant and in the residual solid phase should equal the sum of the amount of cystine added and the amount of cystine in the original urine. Undersaturated urine will dissolve some solid phase cystine, so the amount of solid-phase cystine recovered after 48 h of incubation will be lower than that originally added. In supersaturated urine, cystine precipitates from solution onto added cystine crystals; the amount of solid phase recovered after incubation will be greater than that originally added. Cystine capacity is expressed as the concentration of cystine at saturation, in mg/L.

The primary outcome of the study was the mean cystine capacities in each of the four study periods.

Statistics

All analyses leveraged the within-person dosing schema to focus on the impact of dosing while using each person as his or her own control. Comparisons of outcome variables by dose level were completed by paired t test and with generalized linear models with an identity link and a per-person random effect. We further compared outcome variables to urine sodium (again with generalized linear models and per-person random effect) to assess objectively whether patients were reproducing their dietary regimens. All statistical tests were two-sided and p<0.05 was statistically significant. The analysis was performed using SAS version 9.4 (SAS Institute Inc., Cary NC).

Results

7 patients took tiopronin and 3 patients took d-penicillamine. There were 7 men and 3 women. The patients had an average age of 49 years.

On average, patients had more positive cystine capacity values when they took 1g per day of CBTD compared to 0g (p=0.009, Figure 1 and Table 1). The 24 hour cystine excretion was significantly lower at that dose compared with 0g (P<0.39). However, there was no further change in cystine capacity and no reduction in 24 hour cystine excretion with doses greater than 1g/day. Figures 2A and 2B demonstrate the individual participants’ responses to the incremental dosing regimen. Of the ten participants, all ten had an increase in cystine capacity when the dose was increased from 0 to 1g per day (Figure 2A). Increases in dose from 1 to 2 to 3 g/day caused no further consistent or significant increase in cystine capacity, as clearly shown by the median values in Figure 1. The increase in dose led to a reduction in 24 hour cystine excretion in nine of the 10 participants. Again, increases in dose from 1 to 2 to 3 g/day caused no consistent or significant decrease in cystine excretion. In addition, univariate generalized least squares regression demonstrated that urine volume, phosphorus, and creatinine were predictors of cystine capacity (Table 2).

Figure 1.

Effect of increasing doses of cystine binding thiol drugs on cystine capacity. The dark lines in the boxes are the median values, the length of the box is the interquartile range (IQR) and the whiskers extend to the furthest observation inside of 1.5*IQR from the 25th and 75th percentiles (representing observations outside of the IQR that are not considered outliers). Dots are outliers. The effect of all doses compared with 0 g/day was statistically significant, but the effect of 2 and 3g/day were not different than 1 g/day.

Table 1.

Summary of 24 hour urine measures across CBTD doses (N=10)

	0 g/day Median (IQR)	1 g/day Median (IQR)	2 g/day Median (IQR)	3 g/day Median (IQR)
Volume (L)	2.96 (2.50, 4.09)	3.55 (2.69, 4.72)	2.93 (2.50, 4.72)	3.43 (2.41, 4.50)
Calcium (mg)	124.9 (82.9, 209.0)	114.5 (90.4, 227.0)	136.2 (119.2, 194.0)	153.5 (138.0, 198.1)
Citrate (mg)	816.1 (659.0, 1157.7)	935.9 (747.8, 1310.0)	870.0 (721.7, 1240.9)	1000.9 (791.8, 1168.0)
pH	7.09 (6.76, 7.37)	7.37 (7.08, 7.49)	7.01 (6.87, 7.40)	7.25 (6.66, 7.41)
Sodium (meq)	200.9 (172.5, 253.0)	164.9 (146.6, 230.0)	170.1 (155.6, 218.6)	198.8 (161.4, 288.5)
Creatinine (mg)	2051.2 (1238.0, 2244.1)	2044.1 (1578.7, 2564.0)	1955.2 (1597.4, 2424.3)	2130.8 (1587.2, 2346.0)
Cystine (mg)	1003.9 (699.9, 1146.7)	834.8 (590.3, 1025.0)	650.0 (604.7, 867.0)	870.5 (498.6, 1023.0)
SSCys	1.11 (0.79, 1.20)	0.62 (0.46, 0.87)	0.69 (0.53, 0.93)	0.55 (0.46, 0.99)
Cystine Capacity (mg/L)	−39.1 (−92.0, 48.0)	130.4 (33.4, 168.4)	117.7 (26.4, 149.4)	125.2 (3.4, 242.0)

Abbreviations: IQR=Interquartile Range, SSCys= supersaturation of cystine. Compared with 0g/day, the dose of 1g/day increased cystine capacity (P<0.009) and decreased cystine excretion (P=0.039)

Figure 2A and B.

Individual responses of all participants to increasing doses of CBTDs. A: Cystine capacity. B: 24 hour cystine excretion.

Table 2.

Univariate associations of CBTD and 24h urine variables with the outcome of cystine capacity

CBTH dose	Estimate (95% CI)	p-value
0 grams/day	reference
1 gram/day	172.9 (46.1, 299.7)	0.009
2 grams/day	154.6 (27.8, 281.4)	0.019
3 grams/day	148.6 (21.7, 275.4)	0.023
Volume (per 0.5 L/d)	24.7 (5.9, 43.4)	0.012
Calcium (per 10 mg/d)	−1.86 (−7.64, 3.91)	0.51
Citrate (per 100 mg/d)	3.9 (−10.2, 18.0)	0.58
pH (per 0.1 units)	10.0 (−3.1, 23.1)	0.13
Sodium (per 10 meq/d)	−0.97 (−8.10, 6.17)	0.78
Creatinine (per 100 mg)	−7.93 (−14.03, −1.83)	0.013
Cystine (per 100 mg)	−32.8 (−45.9, −19.7)	<0.001
SSCys (per 0.1 units)	−41.7 (−46.4, −36.9)	<0.001

Abbreviations: CI=Confidence Interval, SSCys=supersaturation of cystine

The p-values are from generalized linear models with a per-person random effect.

Figure 2A also shows that those patients with the most negative cystine capacity values had the greatest positive effect when taking 1g/day compared with 0g/day. Although reproducibility of diet and collections was not ideally maintained throughout the periods, (based on 24h urine sodium, creatinine, not shown), urine sodium was not a significant predictor of cystine capacity.

Changing the dose of the medications was not associated with any adverse effects. No patients experienced any changes in liver enzymes, blood counts or urine protein excretion.

Discussion

A dose of 1g/day of CBTDs led to a significant increase in cystine capacity, representing what is most likely a clinically meaningful increase in cystine solubility. Prescribing doses of CBTDs greater than 1g/day did not significantly further increase cystine solubility, as measured by cystine capacity. In addition, while taking 1g/day of CBTD led to a significant reduction in 24 hour cystine excretion compared with 0g/d in 9 of 10 people, no further reduction occurred with higher doses. It is therefore possible that higher doses of CBTDs may not have clinical benefit beyond a dose of 1g/day. Prescribing the minimum effective dose based on cystine capacity could potentially decrease the adverse effects often associated with thiol drugs.

In our previous study, we demonstrated an increase in cystine capacity when patients went from taking no drug to whatever dose of CBTD they had previously been prescribed.[6] There was a wide range of doses in that earlier study, and given the sample size, no ability to demonstrate whether higher doses were associated with greater effects.

We expected that doses higher than 1g/day would have further effects on the urine variables of interest and lead to greater cystine solubility. We can only speculate on the reasons why higher doses did not lead to augmented effects on urinary cystine capacity. It is possible, but not known, if absorption of these drugs would be limited because of the genetic defect in intestinal cystine transport. The small intestine does manifest defective absorption of cystine (and the other amino acids whose transport is affected)[10], but we do not know that phenotype to affect absorption of CBTDs. However, formation of cysteine-drug complexes could form and limit drug absorption. In retrospect, we realized that measurement of urine levels of the two drugs would have been useful in understanding the observation we made, but did not bank urine specimens from the study.

Studies of the pharmacokinetics of the two CBTDs perhaps explain the findings. Intestinal absorption of only 40–70% of an orally administered dose of d-penicillamine occurs rapidly, with wide inter-individual variation.[11] These studies were not performed in people with cystinuria. Disulfides of the drug could also in part explain decreased absorption and might form in greater amounts with higher doses. Formation of cysteine-drug complexes could also form and impair drug absorption. Absorption may be affected by food, antacids and iron supplements, which we did not control. Similarly, the cumulative urinary excretion of tiopronin is only 34% and highly variable.[12,13] While the maximal concentration is achieved in a relatively short time, justifying at least twice-a-day dosing, there is also a longer in vivo half-life related to plasma protein- and tissue-binding.

We are not aware of any clinical studies demonstrating that increasing doses of these drugs are associated with additional clinical benefit, as judged by reductions in cystine stone burden or activity. The package inserts for both drugs allows for a wide range of dosing. For penicillamine, the recommended dose is 250–1000 mg every 6 hours. Whether such higher doses are associated with greater efficacy in prevention of recurrent stones has not been demonstrated. For tiopronin, the package insert suggests starting at 800 mg/day divided into 3 doses, and then adjusting upward to achieve a cystine concentration of 250 mg/L, without specifying an upper limit to the total daily dose. Our selection of doses was broadly consistent with the package inserts.

We are also not aware of prior studies demonstrating that higher doses of CBTDs lead to lower urinary concentrations or excretion rates of cystine, or if higher doses have little effect as we demonstrate here. We have raised doubt that cystine excretion can be accurately measured by a colorimetric assay technique in the presence of CBTDs, one of the rationales for developing the solid phase assay used here.[4] Measurements of urinary cystine as one of the different sulfhydryl-containing species, including drug-cysteine and drug-drug disulfides in the urine may not be completely accurate.[13]

This relatively small study has several limitations. As cystinuria is a rare disease, the number of patients available and eligible to participate in this study is necessarily small. No randomized controlled trials have been performed to demonstrate the efficacy of different drug doses or other therapeutic strategies, leaving us only with case series and anecdotal reports. It would be challenging to achieve a projected sample size were such a study to be designed. Instead, we use the surrogate outcome of urinary cystine capacity with the hope that it is a satisfactory correlate of kidney stone formation. In our recent study, not yet published, more positive cystine capacity values were associated with a modest statistically significant reduction in stone-related events. However, this effect was not demonstrated in another, recently published retrospective study.[14] We are confident that 1g of CBTD has an important effect on cystine capacity but a lack of benefit of higher doses may not be ruled out due to the sample size.

Another limitation of the study is that diet was not perfectly replicated by study participants in each of the four periods. Having participants replicate self-selected diets has usually been more successful in past studies, judged by near-equivalence of 24 hour urine sodium and urea excretion.[15,16] We do not think that the variation in diet and resulting urinary solute excretion rates has an influence on our result. First, the sequence in which the four periods was performed was randomly assigned and so no consistent effect of the urine analytes with changing doses is evident. Second, the least squares regression did not demonstrate any effect of any other urinary variable on cystine capacity. Finally, the consistent effect of the 1g/day dose to improve cystine solubility is not consistent with some cryptic effect of dietary adherence. We also note that urine pH varied to a degree but in a relatively narrow range, consistent with relatively good adherence to prescribed alkali doses; therefore, variation in urine pH, an important determinant of cystine solubility, was not responsible for the effect on cystine capacity. There was no apparent difference between the effects of tiopronin and d-penicillamine, though the numbers of participants taking the two drugs was not sufficient to test for statistical significance.

In conclusion, we found that 1g of tiopronin or d-penicillamine were both associated with increases in cystine capacity, as expected. There was also a reduction in 24 hour cystine excretion. The effect on cystine capacity should be associated with reductions in cystine stone incidence and growth. Higher doses did not lead to further increases in cystine capacity or decreases in cystine excretion. Limiting prescriptions of these thiol drugs to 1g/day may be associated with fewer adverse effects without sacrificing a benefit of higher doses. However, trials with different doses with stone activity as an outcome would be desirable.