| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Department of Pediatric Biochemistry, Royal Hospital for Sick Children (P.M.C.); and the Department of Child Life and Health, University of Edinburgh (P.M.C., S.F.A., J.C.W., R.S., C.J.H.K., W.H.B.W.), Edinburgh, Scotland; and University Children’s Hospital (M.W.E., M.B.R.), Tubingen, Germany
Address all correspondence and requests for reprints to: Dr. Patricia Crofton, Department of Pediatric Biochemistry, Sciennes Road, Edinburgh, Scotland EH9 1LF.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
At diagnosis, mean SD scores were: bone ALP, -1.84; PICP
-1.77; pyridinoline, -1.42; deoxypyridinoline, -1.66; ICTP, -0.42;
P3NP, +1.45; GH, +24.4; IGF-I, -1.70; IGFBP-3, -0.88; IGFBP-2, +2.42;
and GH-binding protein, -0.69. Bone ALP, PICP, and IGFBP-3 were all
correlated (P 
0.03). During induction and
intensification, there was shrinkage of the lower leg, with decreases
in PICP, pyridinoline, ICTP, and P3NP (P < 0.05),
whereas IGF-I and IGFBP-3 increased (P < 0.05).
After prednisolone was discontinued, bone ALP and collagen markers
increased markedly (P < 0.01), but there was no
significant change in IGF-I and IGFBP-3. In 12 children who received
high dose iv methotrexate, postglucocorticoid increases in bone ALP and
PICP were less, whereas those in ICTP and P3NP were greater, compared
to levels in children who did not receive methotrexate
(P < 0.05).
We conclude that ALL itself caused GH resistance and low bone turnover. During early intensive chemotherapy, further suppression of osteoblast proliferation and osteoclast activity occurred, not mediated through the systemic GH axis, probably by the direct action of prednisolone on bone. The postglucocorticoid increase in bone turnover was also independent of the GH axis and was modulated by high dose iv methotrexate, which depressed osteoblast recovery and enhanced osteoclast activity.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Impaired growth and poor bone quality may be caused by a number of factors, including the disease process itself, concurrent serious infections, poor nutrition, and inactivity, in addition to the various components of chemotherapy. Changes in height SD scores and bone mineral density cannot be reliably assessed over periods of less than 6 months and, therefore, cannot themselves provide a sufficiently detailed reflection of growth and bone turnover to allow conclusions to be reached as to possible causative factors. We have previously reported that lower leg length velocity (LLLV) is decreased during the first 6 weeks of intensive chemotherapy for ALL, but increases thereafter (14). We report here the results of a randomized, longitudinal, prospective study in which we made serial measurements of a panel of suitably chosen biochemical markers to provide a dynamic reflection of whole body growth and bone turnover associated with each phase of chemotherapy over the first 6 months of treatment. This has not only clarified which phases of chemotherapy contribute most to any growth deficit and detrimental effects on bone, but has also provided insight into the mechanisms involved.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We studied 22 children with ALL during the first 6 months of chemotherapy. This group of children was part of the Medical Research Council-funded randomized trial of childhood ALL treatment in the UK, which was performed from 1992 to 1997 (UKALL XI-92). All children presenting to the Royal Hospital for Sick Children (Edinburgh, Scotland) between January 1993 and June 1994 with a diagnosis of ALL were eligible to participate. The median age of the children was 4.4 yr (range, 1.2–13.5 yr). Seventeen boys and five girls enrolled in the study. Two boys were in early puberty at diagnosis, but showed no pubertal progression during treatment. The remaining children remained prepubertal throughout the study. Median white cell count at diagnosis was 9.3 x 109/L (range, 1.3–435,000 x 109/L). All had entered clinical remission by week 5 of induction chemotherapy. Three patients relapsed later during treatment, at 35 weeks, 46 weeks, and 18 months, respectively. As central nervous system (CNS)-directed treatment, two patients received cranial irradiation, and the remainder were randomized to receive either continuing intrathecal methotrexate (n = 8) or high dose iv methotrexate plus intrathecal methotrexate (n = 12). The study was approved by the local ethics committee, and informed consent was obtained from parents and (where appropriate) children.
Samples and anthropometric measurements
For children randomized to high dose iv methotrexate plus
intrathecal methotrexate, blood samples were obtained at the intervals
shown in Fig. 1
. During CNS-directed
treatment, the samples were timed to coincide with vincristine
administration, as this was when children attended the hospital and
when venous access was available. For this reason, blood samples were
collected from children randomized to intrathecal methotrexate alone at
12 and 16 weeks (coinciding with vincristine administration) instead of
14 and 18 weeks as shown in Fig. 1. At all other times, samples were
collected according to an identical schedule in both groups. In five
children, additional blood samples were collected at week 21, on
completion of second intensification; this was not possible in the
remaining children because of lack of venous access. Samples were
collected between 1100–1500 h to minimize the effects of circadian
variation. Plasma and serum were stored in aliquots at -70 C until
analysis.
|
. However, the diuresis
associated with induction chemotherapy and the stress on families
during the period of study made it impossible to obtain a complete
dataset. The volume and creatinine concentration of each urine
collection were measured immediately. Aliquots of urine were then
stored at -70 C until analysis for pyridinium cross-links (thiomersal
preservative only) and GH (stabilized with 1 g/L BSA).
Height was measured in all patients by a trained operator using a
Holtain stadiometer, at diagnosis and during weeks 5–7, 18–20, and
23–27, respectively. We measured lower leg length by knemometry, in
the morning, using the random zero method (15); measurements were made
at the intervals shown in Fig. 1 in those children who were ambulant
and old enough to cooperate. The median technical error (1
SD from the mean of a set of triplicate measurements) was
0.15 mm.
Analytical methods
Collagen assays. We measured procollagen type I C-terminal propeptide (PICP), the cross-linked telopeptide of type I collagen (ICTP), and procollagen type III N-terminal propeptide (P3NP) in plasma by RIA (Orion Diagnostica, Espoo, Finland), using methods previously described (16, 17, 18). Before analysis, we diluted samples appropriately in 154 mmol/L sodium chloride to achieve concentrations within the calibration curve; typical dilutions were 1:4 for PICP and 1:2 for ICTP and P3NP. All samples were analyzed in duplicate. As far as possible, samples from each patient were analyzed in a single analytical run to minimize analytical variation. Between-run coefficients of variation were 7.8% and 5.2% at 94 and 320 µg/L for PICP, 6.3% and 9.2% at 8.7 and 33.8 µg/L for ICTP, and 5.6% and 6.4% at 4.6 and 10.4 µg/L for P3NP.
Bone alkaline phosphatase (ALP). Bone ALP was measured in plasma by wheat-germ lectin affinity electrophoresis, as previously described (19). Between-run coefficients of variation were 2.2%, 3.5%, and 1.9% at 251, 349, and 435 U/L, respectively.
Pyridinium cross-links. Pyridinium cross-links were measured in urine by high performance liquid chromatography using a modification of the method of Pratt et al. (20). In urine, approximately 40% of pyridinium cross-links are in the unconjugated (free) form, and the remainder are either glycosylated or present as low mol wt peptides. The urine samples were hydrolyzed to convert the glycosylated and peptide forms to the free form and, after solid phase extraction using CC31 microgranular cellulose, total pyridinoline (Pyd) and deoxypyridinoline (Dpd) were measured by isocratic reverse phase high performance liquid chromatography, using a C8 (4.6 x 250-mm) column, heptafluorobutyric acid as the ion-pairing agent, and acetonitrile as the organic modifier in the mobile phase. Pyridinium cross-links extracted from demineralized sheep bone were used as external standards (21) and were a generous gift from Dr. Simon Robins (Rowett Research Institute, Aberdeen, Scotland). The results were expressed in relation to creatinine measured on the same urine sample. Samples were assayed in duplicate with mean interassay coefficients of variation of 5.9% at 315 nmol/mmol creatinine for Pyd and 9.1% at 65.8 nmol/mmol creatinine for Dpd. The interassay coefficient of variation for the ratio of Pyd to Dpd was 7.9% at a ratio of 4.8.
Insulin-like growth factor I (IGF-I). IGF-I was measured in serum with a specific RIA (Mediagnost, Tubingen, Germany). This assay uses an excess of IGF-II to eliminate interferences by IGF-binding proteins (IGFBPs) (22). Between-assay coefficients of variation were 8.5%, 6.5%, and 8.0% at 69, 140, and 118 µg/L, respectively.
IGFBP-3. IGFBP-3 was measured in serum using a specific RIA as previously described (23). Between-assay coefficients of variation were 7.3% and 6.9% at 2772 and 3545 µg/L, respectively.
IGFBP-2. IGFBP-2 was measured in serum using a specific RIA (24). Recombinant human IGFBP-2 (a gift from Sandoz, Basel, Switzerland) was used as a standard and as the tracer. The sensitivity of the assay is 0.2 µg/L. The between-assay coefficient of variation was 10.7%.
GH. GH was measured in urine by an amplified enzyme immunoassay (Novoclone, NovoNordisk, Crawley, UK). The results were expressed in relation to creatinine measured on the same urine sample. Between-run coefficients of variation were 8.0%, 13.9%, and 15.2% at 37.2, 9.2, and 4.7 ng/L, respectively.
GH-binding protein (GHBP). GHBP was measured by an in-house RIA following the principles published by Kratzsch et al. (25). Radiolabeled (125I) recombinant human (rh) GHBP was a gift from J. Kratzsch (Leipzig, Germany). Antibody against rhGHBP (batch 1070, Pharmacia & Upjohn, Stockholm, Sweden) was raised in rabbits. This material was also used as standard. The sensitivity of the assay was 0.02 ng/mL. The between-assay coefficient of variation was 8.3% at 1.5 ng/mL.
Data analysis
Height measurements were expressed as age- and sex-specific SD scores in relation to published data (26). The distributions of serum concentrations of collagen markers, IGF-I, and IGFBPs in healthy children are log-normal (27, 28). We therefore transformed measured concentrations to their logarithms before calculating age- and sex-specific SD scores based on our own published data (27, 28). Bone ALP and urinary Pyd, Dpd, and GH did not require log transformation. For bone ALP, SD scores were calculated in relation to our own published data (29). For urinary Pyd and Dpd, we calculated SD scores in relation to urine samples from 14 healthy children, aged 2–13 yr, in our own local population (unpublished data). For urinary GH, SD scores were calculated in relation to 12-h overnight urine collections from 53 healthy children in our local population (unpublished data). For GHBP, SD scores were calculated based on normative data from 241 healthy children, aged 1–20 yr (unpublished data). For the knemometry data, we calculated LLLV for each time point by subtracting the lower leg length at that time point from the length measured at the previous time point, and dividing by the exact time interval between the two measurements. Results were expressed as millimeters per week.
We included data from the two children who received cranial irradiation up to the time point immediately preceding irradiation; all subsequent results from those children were excluded from analysis. We also separately analyzed data from the five children from whom we obtained samples at week 21 to examine the effects of a second intensification in greater detail. The mean and 95% confidence limits of the mean were calculated for each analyte at each time point. We employed nonparametric statistical tests throughout because variances were not always equal despite log transformation. Data were compared longitudinally through time using Wilcoxon signed rank tests; comparisons between groups were by Mann-Whitney U tests with correction for ties. Spearman rank correlations with correction for ties were used to compare variables at each time point. All statistical tests were two-tailed; P < 0.05 was regarded as significant.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
At diagnosis, mean height SD score was +0.38 (95%
confidence interval, -0.03 to +0.79). Markers of bone formation (bone
ALP and PICP) and degradation (ICTP, Pyd, and Dpd) were all low
compared to those in the healthy population (Fig. 2, a, b, and d–f). By contrast, P3NP was
mildly increased (Fig. 2c). The median Pyd/Dpd ratio at diagnosis was
9.5 (range, 1.5–68.3) compared to a mean ratio of 5.0 (range,
3.3–6.6) in healthy children. Urinary GH excretion was markedly
increased at diagnosis, but IGF-I and IGFBP-3 were decreased, whereas
IGFBP-2 was increased (Fig. 2, g–i). The mean GHBP SD
score was -0.69 (95% confidence interval, -1.17 to -0.21).
|
Induction of remission (weeks 1–5)
All children entered clinical remission from their disease during
this period. During induction, there was a negative LLLV, indicating
shrinkage of the lower leg (Fig. 2j). Markers of collagen synthesis,
PICP and P3NP, and collagen degradation, Pyd, all decreased
significantly. However, mild decreases in ICTP and Dpd did not reach
statistical significance (P > 0.34). Bone ALP
initially increased at week 3, then decreased again by week 5. There
was little change in the high excretion of urinary GH during this
period, but IGF-I and IGFBP-3 increased and IGFBP-2 decreased to
concentrations matching those in the normal population.
First intensification (weeks 5–6)
By the end of first intensification, the mean height
SD score had decreased by -0.31 compared to the mean
pretreatment SD score (95% confidence limits, -0.42 to
-0.20) and the lower leg maintained a steady negative growth velocity
(Fig. 2j). There were further decreases in PICP and P3NP and lower
concentrations of ICTP compared to pretreatment levels, but no
statistically significant change in Pyd and Dpd excretion. Urinary GH
began to decrease at this stage, but there was no further change in
IGF-I, IGFBP-3, or IGFBP-2 concentrations.
CNS-directed treatment and continuing chemotherapy (weeks 6–20)
After withdrawal of prednisolone at the end of
first intensification, there were dramatic increases in bone ALP, PICP,
ICTP, and P3NP (Fig. 2). ICTP reached its peak earliest at week 8, 2
weeks after completion of the first intensification, then gradually
decreased. PICP and P3NP also showed an early rise in concentration,
but continued to increase further to peak at around 12–16 weeks. Bone
ALP was slower to increase, but peaked at the same time as PICP,
although, unlike PICP, it remained low relative to that in the healthy
childhood population (mean peak SD score -0.63). Changes
in Pyd and Dpd were more variable, and the data were too few to reach
significance, but the trend in these urinary markers of collagen
breakdown supported those in the plasma marker, ICTP. During this
period, there was little change in IGF-I and IGFBP-3 until a small
increase occurred around week 20. However, IGFBP-2 increased to peak at
weeks 16–18, then decreased markedly at week 20. LLLV showed a
significant increase at weeks 12–14, coinciding with the period of
increasing PICP and bone ALP and decreasing ICTP and Pyd. This net
growth of the lower leg was subsequently maintained at a constant
rate.
Children who were randomized to receive high dose iv methotrexate
showed no significant difference in any of the markers, up to and
including week 8 of chemotherapy, compared to children who received
intrathecal methotrexate alone. However, at week 14, the group who had
just completed three courses of high dose iv methotrexate had
significantly lower bone ALP and PICP levels and significantly higher
P3NP and ICTP levels than the group who had received only intrathecal
methotrexate (Fig. 3). IGF-I, IGFBP-3,
and IGFBP-2 did not differ between the two groups at any time.
|
For all five children from whom a sample was obtained at week 21,
PICP, P3NP, and IGFBP-3 decreased, whereas IGFBP-2 increased
between weeks 20 and 21 (P < 0.05; Fig. 4). PICP and ICTP increased again in
every child between weeks 21 and 23 (P < 0.05).
Similar trends were seen for the other markers, but did not reach
statistical significance due to the small numbers of
subjects.
|
). Bone ALP,
PICP, P3NP, IGF-I, IGFBP-3, and IGFBP-2 all subsequently showed
significant increases between weeks 23 and 27 during continuing
chemotherapy. Correlations between markers during chemotherapy
IGF-I showed significant positive correlations with IGFBP-3 and inverse correlations with IGFBP-2 at most time points (P < 0.05), but there was no significant correlation between IGFBP-3 and IGFBP-2 at any time point. There were few other significant correlations among biochemical markers after chemotherapy began, except at week 8, when bone ALP, PICP, and P3NP showed significant positive correlations with one another and with IGF-I (P < 0.05), but inverse correlations with IGFBP-2 (P < 0.03). Among the five children from whom a sample was collected at week 21, IGF-I was positively correlated with PICP (P < 0.05), but no other statistically significant correlations were found at that stage.
Relationship of markers to neutrophil counts during chemotherapy
The period of neutropenia (neutrophil count <1.0 x 109/L) and subsequent recovery after second intensification coincided with a similar pattern in bone ALP, PICP, P3NP, IGF-I, and IGFBP-3. However, there were no significant correlations between neutrophil counts and any marker at any time point, except for an inverse correlation with PICP at week 23 (P = 0.0006) and a positive correlation with IGFBP-3 at week 27 (P = 0.05).
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Sporadic reports of bone pain, fracture, and osteoporosis in some children treated with methotrexate on a long term basis (31, 32, 33, 34) have led to speculation that methotrexate may be responsible for adverse effects on bone in more recent ALL chemotherapy protocols, but there has been no direct evidence for this. Two studies in children with ALL have suggested that bone mineral content may be low at diagnosis and decrease further during chemotherapy (12, 13), but the cause was unclear. We have used a panel of biochemical markers reflecting different aspects of bone and collagen turnover to clarify the relative effects of disease and chemotherapy together with the mechanisms involved.
PICP and bone ALP were chosen to reflect different stages of osteoblast function; ICTP, Pyd, and Dpd were chosen as markers of collagen degradation, largely (but not exclusively) arising from bone (35); and P3NP was chosen as a marker of soft tissue collagen synthesis. We measured GH in overnight timed urinary collections to provide an integrated noninvasive reflection of GH secretion but recognized that renal handling was likely to be disturbed during fluid loading and diuresis immediately before and during induction chemotherapy for ALL and also by the drugs administered. Expressing excretion in relation to creatinine only partly compensated for this because the children were in a catabolic state that was likely to affect creatinine. Urinary GH results in this study should therefore be interpreted with caution. To provide additional information, we measured serum IGF-I, IGFBP-3, and IGFBP-2. IGFBP-2 has been less extensively investigated than either IGF-I and IGFBP-3, but appears to be inversely related to GH secretion (36) and inhibits IGF-I stimulated bone cell proliferation and bone collagen synthesis (37, 38). To assist in the interpretation of these markers of the GH axis, we also measured circulating GHBP at diagnosis.
At diagnosis (pretreatment)
We, like others (1, 4, 5, 9, 30), found no apparent impairment of height in children with ALL at diagnosis when compared cross-sectionally with the healthy population. However, one longitudinal study described decreasing SD scores for height and weight for at least 1 yr before diagnosis, although all scores remained greater than zero (39). Our data are in agreement with these observations and demonstrate clearly that at diagnosis children with ALL were in a low bone turnover state caused by the disease itself. A similar low bone turnover state has been described by Sorva et al. (40). Both IGF-I and IGFBP-3 were low at diagnosis, in agreement with a previous report (41); GHBP levels were also relatively low, and urinary GH appeared to be very high, suggesting that these children were in a GH-resistant state, similar to that described in septicemia (42). Recent poor nutrition may also have contributed to low IGF-I, IGFBP-3, and GHBP and high IGFBP-2 levels. GHBP was positively correlated with Dpd, and we observed trends toward positive relationships with bone ALP, IGF-I, and IGFBP-3 (P < 0.15) that did not reach statistical significance but were consistent with the concept that down-regulation of the number of GH receptors at least partially contributed to the GH-resistant state and reduced bone turnover. The strong correlations among IGFBP-3, PICP, and bone ALP suggest that GH resistance was causally related to reduced bone formation.
By contrast, IGFBP-2 was high at diagnosis, confirming previous reports that it is increased in ALL (41, 43). Its inverse correlation with IGF-I, contrasted with the positive correlation between IGFBP-3 and IGF-I and the lack of correlation between IGFBP-2 and IGFBP-3, suggest that both BPs may have independent and opposing regulatory roles in controlling IGF-I bioavailability. These correlations persisted through most of the chemotherapy period (data not shown).
Unlike the other collagen markers, P3NP was increased at diagnosis, with higher levels in those children who subsequently relapsed during chemotherapy. It seems unlikely that P3NP was reflecting overall somatic growth at a time when bone growth and turnover were clearly suppressed. P3NP may instead have been reflecting tumor burden, possibly derived from leukemic masses. The high Pyd/Dpd ratio at diagnosis compared to that in normal children may also have reflected the increased turnover of such soft tissue masses (exclusively Pyd cross-links) compared to the low bone turnover (Dpd plus Pyd cross-links). P3NP has previously been shown to have prognostic value in adult patients with soft tissue sarcomas (44).
Induction and first intensification
We demonstrated progressive suppression of all collagen markers of bone and soft tissue turnover, height loss, and shrinkage of the lower leg during this period. The bone formation marker, PICP, was more affected than markers of bone resorption, ICTP and Dpd. We have previously reported decreased height, sitting height, and sitting height/height ratio and increased weight over the same period (10). The loss of height, sitting height, and lower leg length without any biochemical evidence of increased bone resorption may be at least partly explained by compression of intervertebral tissue and associated loss of elasticity in the spine together with compression of overlying soft tissue in the lower leg as a result of weight gain (45).
Prednisolone was the most likely agent to have caused the observed biochemical effects. Glucocorticoids inhibit collagen biosynthesis and down-regulate the proliferation of both osteoblasts and chondrocytes in the epiphyseal growth plate (6, 46, 47, 48, 49). Studies in children with asthma and inflammatory bowel disorders have confirmed that both PICP and ICTP decrease after commencing glucocorticoid treatment, in agreement with our observations (50, 51). In our study, the apparent paradoxical increase in bone ALP 2 weeks after commencing prednisolone was probably caused by down-regulation of osteoblast proliferation by glucocorticoids, resulting in induction of synthesis of ALP as osteoblasts prematurely entered their maturation phase (46, 47, 52). The lack of subsequent recruitment into the maturation phase from the depleted population of osteoblasts and their precursors would result in the later decrease in bone ALP during weeks 5–6.
We observed an increase in serum IGF-I and IGFBP-3 and a decrease in IGFBP-2 to normal levels during induction. Improved nutrition and increased well-being associated with induction of clinical remission may have contributed to these changes, but similar changes in circulating levels have also been reported in adult volunteers treated with dexamethasone (53). However, circulating IGF-I may not reflect tissue concentrations; a reduction of IGF-I at the cellular level by prednisolone would result in the changes in bone turnover markers observed in our patients, mediated by paracrine and autocrine rather than endocrine effects. The demonstration of a direct inhibitory effect of dexamethasone on the epiphyseal growth plate in rabbits (54) also suggests that prednisolone was largely responsible for the cessation of growth in our patients through its effects on local production of IGF-I in the proliferative zone of the epiphyseal growth plate (55).
After first intensification and during CNS-directed treatment (weeks 6–20)
After cessation of prednisolone treatment at the end of first intensification, there was a rapid and dramatic increase in all markers of collagen turnover, coinciding with a resumption of linear growth. Bone ALP increased more slowly than PICP, reflecting its later synthesis in the osteoblast cycle and, unlike PICP, remained well below the population mean. As a considerable proportion of newly synthesized type I collagen (as reflected by PICP) is degraded without ever being laid down into bone, bone ALP, being both bone specific and necessary for mineralization of the newly formed collagen matrix, is likely to represent more accurately the amount of mineralized bone formed during this period of recovery from the combined effects of leukemia and prednisolone. At the same time, urinary GH decreased, but there was little change in mean circulating levels of IGF-I and IGFBP-3, suggesting that both suppression and recovery of growth and bone turnover occurred largely independently of the GH axis. However, it is likely that locally produced IGF-I or other growth factors were involved at the tissue level in mediating these responses through paracrine or autocrine effects.
Our data indicate that patients given high dose iv methotrexate in addition to intrathecal methotrexate had lower bone ALP and PICP but higher P3NP and ICTP than patients given only intrathecal methotrexate. In adults treated with methotrexate for psoriasis, increased P3NP reflected liver fibrosis or cirrhosis (56). However, there have been no previous studies of collagen markers or bone ALP in the systemic circulation in children treated with methotrexate. In our study, the higher levels of plasma P3NP found in the children immediately after high dose iv methotrexate therapy compared to those in the group treated only with intrathecal methotrexate may have been due to occult liver fibrosis, which has been reported as a side-effect of methotrexate administration (32). We found no difference between the two groups in liver ALP or in alanine transaminase (data not shown), suggesting that P3NP may be a more sensitive indicator of this often insidious process than conventional liver function tests.
Our results for PICP, bone ALP, and ICTP suggest that high dose iv methotrexate depressed bone formation and enhanced bone resorption during this crucial period of recovery from a prolonged state of low bone turnover. In histomorphometry studies with adult rats, methotrexate has been shown to decrease trabecular bone volume, bone mass, bone formation, and osteoblast activity and increase osteoclast activity (57, 58). ALP and osteocalcin were lower, and hydroxyproline (a marker of bone resorption) was higher in methotrexate-treated rats than in controls (57), analogous to our findings in children. With regard to growth, methotrexate has no apparent inhibitory effect on uptake of [35S]sulfate or [3H]thymidine by porcine costal cartilage (6), little effect on short term tibial growth in rabbits (59), and no discernible effect on height velocity in children with ALL (10). The effects of high dose iv methotrexate are therefore more likely to be on bone architecture and quality than on longitudinal growth.
Second intensification (week 20)
The greater quantitative effects on biochemical markers of second compared to first intensification may be ascribed to much higher preintensification marker concentrations. In contrast to first intensification, IGF-I and IGFBP-3 decreased significantly, and IGF-I was correlated with PICP, suggesting that at least some of the deleterious effects on bone and soft tissue synthesis were mediated through decreased circulating IGF-I.
In summary, we have demonstrated that at diagnosis, children with ALL were in a state of low bone turnover, probably caused by GH resistance associated with the disease itself. During induction and first intensification, there was further height loss, shrinkage of the lower leg, and suppression of markers of bone and soft tissue turnover. These effects occurred independently of circulating IGF-I and IGBP-3 and were likely to be the result of a direct effect of prednisolone on target tissues. Once prednisolone was discontinued, all markers of bone and soft tissue turnover increased dramatically, again independently of the GH axis, although bone ALP did not reach the population mean, suggesting suboptimal bone formation. Children treated with high dose iv methotrexate showed evidence of poorer bone formation and increased bone resorption than those who received only intrathecal methotrexate. This study has clarified the mechanisms by which both the disease process and intensive chemotherapy may affect growth and bone quality in children with ALL and indicates the need for long term follow-up to assess the risks of later osteoporosis.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received February 6, 1998.
Revised June 2, 1998.
Accepted June 16, 1998.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. K Vihinen, K.-L. Kolho, M. Ashorn, M. Verkasalo, and T. Raivio Bone turnover and metabolism in paediatric patients with inflammatory bowel disease treated with systemic glucocorticoids Eur. J. Endocrinol., December 1, 2008; 159(6): 693 - 698. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Ruble Skeletal Complications After Bone Marrow Transplant in Childhood Journal of Pediatric Oncology Nursing, April 1, 2008; 25(2): 79 - 85. [Abstract] [PDF]
|
|
|
|
 |
|
 D. Marinovic, S. Dorgeret, B. Lescoeur, C. Alberti, M. Noel, P. Czernichow, G. Sebag, E. Vilmer, and J. Leger Improvement in Bone Mineral Density and Body Composition in Survivors of Childhood Acute Lymphoblastic Leukemia: A 1-Year Prospective Study Pediatrics, July 1, 2005; 116(1): e102 - e108. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Wiernikowski, R. Barr, C Webber, C. Guo, M Wright, and S. Atkinson Alendronate for steroid-induced osteopenia in children with acute lymphoblastic leukaemia or non-Hodgkin's lymphoma: results of a pilot study Journal of Oncology Pharmacy Practice, June 1, 2005; 11(2): 51 - 56. [Abstract] [PDF]
|
|
|
|
 |
|
 B. M. D. Brennan, Z. Mughal, S. A. Roberts, K. Ward, S. M. Shalet, T. O. B. Eden, A. M. Will, R. F. Stevens, and J. E. Adams Bone Mineral Density in Childhood Survivors of Acute Lymphoblastic Leukemia Treated without Cranial Irradiation J. Clin. Endocrinol. Metab., February 1, 2005; 90(2): 689 - 694. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. M. Crofton, N. Evans, M. R.H. Taylor, and C. V. Holland Procollagen Type I Amino-Terminal Propeptide: Pediatric Reference Data and Relationship with Procollagen Type I Carboxyl-Terminal Propeptide Clin. Chem., November 1, 2004; 50(11): 2173 - 2176. [Full Text] [PDF]
|
|
|
|
 |
|
 K. J. Elliot, S. J. Millward-Sadler, M. O. Wright, J. E. Robb, W. H. B. Wallace, and D. M. Salter Effects of methotrexate on human bone cell responses to mechanical stimulation Rheumatology, October 1, 2004; 43(10): 1226 - 1231. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Mushtaq, P. Bijman, S. F. Ahmed, and C. Farquharson Insulin-Like Growth Factor-I Augments Chondrocyte Hypertrophy and Reverses Glucocorticoid-Mediated Growth Retardation in Fetal Mice Metatarsal Cultures Endocrinology, May 1, 2004; 145(5): 2478 - 2486. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Mayo, H. Macintyre, A. M. Wallace, and S. F. Ahmed Transdermal Testosterone Application: Pharmacokinetics and Effects on Pubertal Status, Short-Term Growth, and Bone Turnover J. Clin. Endocrinol. Metab., February 1, 2004; 89(2): 681 - 687. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. H. Davies, B. A. J. Evans, M. E. M. Jenney, and J. W. Gregory Effects of Chemotherapeutic Agents on the Function of Primary Human Osteoblast-Like Cells Derived from Children J. Clin. Endocrinol. Metab., December 1, 2003; 88(12): 6088 - 6097. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. M. Crofton, N. Evans, and R. Stephen Serum CrossLaps Compared with Other Markers of Bone Turnover in Severely Malnourished Children before and after Refeeding Clin. Chem., January 1, 2003; 49(1): 192 - 195. [Full Text] [PDF]
|
|
|
|
 |
|
 T Mushtaq and S F Ahmed The impact of corticosteroids on growth and bone health Arch. Dis. Child., August 1, 2002; 87(2): 93 - 96. [Full Text] [PDF]
|
|
|
|
 |
|
 D. Kinniburgh, H. Zhu, L. Cheng, A.T. Kicman, D.T. Baird, and R.A. Anderson Oral desogestrel with testosterone pellets induces consistent suppression of spermatogenesis to azoospermia in both Caucasian and Chinese men Hum. Reprod., June 1, 2002; 17(6): 1490 - 1501. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. W. Vesper, L. M. Demers, R. Eastell, P. Garnero, M. Kleerekoper, S. P. Robins, A. K. Srivastava, G. R. Warnick, N. B. Watts, and G. L. Myers Assessment and Recommendations on Factors Contributing to Preanalytical Variability of Urinary Pyridinoline and Deoxypyridinoline Clin. Chem., February 1, 2002; 48(2): 220 - 235. [Abstract] [Full Text] [PDF]
|
3.0 yr);
high dose iv methotrexate, 6 or 8 mg/m2; oral methotrexate,
20 mg/m2; and mercaptopurine, 75 mg/m2.
