| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology, The University of Melbourne, Royal Melbourne
Hospital, Grattan St., Parkville, Melbourne, Victoria 3050, Australia.
2 Department of Radiology, Academic Medical Centre, University of Amsterdam,
Amsterdam, The Netherlands.
3 School of Paediatrics and Child Health, University of Western Australia,
Perth, WA, Australia.
Received October 23, 2006;
accepted after revision January 26, 2007.
Address correspondence to P. L. Robertson.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Serial MR images of the femurs, lumbar spine, or both of 57 subjects with a diagnosis of Gaucher's disease, 44 of whom were being treated with enzyme replacement therapy, were evaluated by two musculoskeletal radiologists using the bone marrow burden (BMB) scoring system. This system gives a score out of a possible 8 for the lumbar spine and a score out of a possible 8 for the femurs, so the total BMB score is out of 16.
RESULTS. The mean total BMB scores at baseline and final measurement were 13 (95% CI, 12.0–13.8) and 6.5 (95% CI, 5.2–7.9), respectively. Total BMB scores improved by 2–12 points (mean, 6.3 points) for 15 subjects in whom total BMB scores obtained before enzyme replacement therapy and at least one time point after commencement of therapy were available. In 39 subjects for whom only baseline and final femoral BMB scores were available, 24 subjects (62%, 95% CI, 47–76%) improved by 2 or more points while on enzyme replacement therapy. In 24 subjects for whom only baseline and final lumbar BMB scores were available, the BMB score in 16 subjects (67%, 95% CI, 48–85%) improved by 2 or more points.
CONCLUSION. The use of the BMB score enabled semiquantitative assessment of bone marrow response to enzyme replacement therapy in adult patients with type 1 Gaucher's disease in Australia. The ability to assess therapy response will facilitate the tailoring of dosage regimens in the future.
Keywords: bone marrow burden score • enzyme replacement therapy • Gaucher's disease • MRI • musculoskeletal system
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In terms of organ-specific complications, no data were available about the management of skeletal disease with enzyme replacement therapy in terms of optimal dosage regimens and objective parameters of response. In Australia, a national program was developed by the federal health authorities in 1994 to facilitate delivery of enzyme replacement therapy through a centralized coordinated process [2]. This approach not only rationalized the equitable availability of enzyme replacement therapy to all significantly affected individuals but also allowed an ongoing assessment of drug efficacy and safety. The use of standardized monitoring protocols and data capture also generated information relevant to the analysis of various aspects of the disease and its management. Through this process, the clinical efficacy of enzyme replacement therapy on skeletal disease was documented with structured serial MRI measures, with all images being centrally analyzed in a qualitative manner by a single radiologist who was blinded to the other aspects of each patient's clinical disease or therapeutic status.
Several scoring methods have been used to determine the severity of bone disease and assess response to therapy, both quantitative [3–6] and semiquantitative [7–15]. The Dixon quantitative chemical shift imaging (QCSI) method is reported to be the most sensitive quantitative method [16] but is not available at most imaging sites and has not been available in Australia. Recently, a semiquantitative bone marrow burden (BMB) scoring system has been described that correlates with QCSI and has been shown to detect response to therapy. Using this BMB score, radiologists can assess the femoral and axial bone marrow by giving a total BMB score of up to 16 points [15]. This scoring system has also been described more recently in comparing two large European cohorts [17].
The purpose of this study was to evaluate the BMB scoring system for objectively quantifying serial response of bone marrow disease to enzyme replacement therapy in a population of adults with type 1 Gaucher's disease in Australia and thus facilitate tailoring of enzyme replacement therapy dose to skeletal disease burden.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Subjects
Fifty-seven patients with a diagnosis of Gaucher's disease for whom MR
images of the lumbar spine or femurs were available for review were included
in this study. All patients had a confirmed diagnosis of Gaucher's disease on
the basis of ß-glucocerebrosidase enzyme assay or gene mutation analysis.
The patients were assessed for and monitored on enzyme replacement therapy by
the Gaucher Advisory Committee (GAC) of the Australian Federal Department of
Health and Aging according to guidelines previously developed
[18]. The responses of
nonskeletal parameters for 48 of these patients have been reported previously
[2].
The study group included 29 women and 28 men. Ages ranged from 19 to 75 years (mean, 45 years; median, 45 years). Forty-four subjects had been on enzyme replacement therapy for between 1 and 12 years (mean, 3.5 years; median, 4 years) before the final MRI examination. Thirteen subjects had previous splenectomies.
MRI
MRI had been performed at a number of different sites using standardized
guidelines for imaging that included T1- and T2-weighted sequences of the
lumbar spine and coronal T1-weighted and STIR (or fat-suppressed sequences) of
the femurs. However, varying protocols were used, so adequate images were
available for analysis at baseline of the spine in 30 subjects, of the femurs
in 53, and of both the femurs and spine in 25 subjects.
MR images of each subject were viewed as hard copy or on a computer workstation in DICOM format in consensus by two musculoskeletal radiologists experienced in the interpretation of MR images in Gaucher's disease using a BMB scoring system adapted from Maas et al. [15]. The radiologists were blinded to whether the subjects were on enzyme replacement therapy at the time of the examination or at any time before the examination. Images were interpreted in one sitting over 2 days.
In 14 subjects on enzyme replacement therapy, the first MR images available for review had been obtained between 10 and 136 months (mean, 51 months; median, 38 months) after commencement of enzyme replacement therapy. These patients commenced enzyme replacement therapy before their initial MR assessment because of a lack of access to MRI, but it was still considered valid to compare their serial BMB scores while on therapy. Unless specifically stated, "baseline data" refers to the first MR images available whether the images were obtained before or after commencement of enzyme replacement therapy.
BMB Scoring Method: Adapted from Maas et al. [15]
A score was given for each examination of the lumbar spine in which
spin-echo or fast spin-echo non–fat-suppressed T2- and T1-weighted
images were available. Table 1
shows the scoring system for signal intensity in the lumbar spine. A score of
0–3 was given for signal intensity on the T1-weighted images and a score
of 0–2 for signal intensity on the T2-weighted images. A score of
1–3 was given for distribution of bone marrow abnormality depending on
whether it was patchy (score = 1), whether it was diffuse (score = 2), or
whether there was involvement of the fat surrounding the basivertebral veins
(score = 3). Infarcts in the lumbar spine were categorized as small or as
extensive if they involved a whole vertebral body.
|
A signal intensity score was given for images of the femurs if coronal T1- and T2-weighted images (non–fat-suppressed or fat-suppressed) of the length of both femurs were available (Table 2). If non–fat-suppressed images of the femurs were supplied, a score of 1 and 2 was given for T2-weighted images if the marrow was slightly hypointense or definitely hypointense, respectively.
|
The extent of involvement was scored depending on whether the diaphyses and metaphyses were involved alone (score = 1) or whether there was also involvement of the proximal epiphyses (score = 2) or the distal epiphyses (score = 3).
The presence of small medullary or extensive medullary infarcts and infarction involving the epiphyseal sites (avascular necrosis) was recorded separately.
If a joint had been replaced, it was presumed that this had been for previous infarction. It was assumed that the infiltration score was the same as for the same epiphysis in the contralateral limb. However, a score of 2 or 3 was given for infarction depending on which epiphysis was involved.
If there was extensive increased T2 signal in regions that were clearly infarcted, a score of 1 rather than 2 was given for T2 signal intensity. If heterogeneous or mixed signal was clearly due to infarction and noninfarcted marrow was clearly not affected by heterogeneous signal, a score was given for the noninfarcted marrow only.
Total BMB scores for the lumbar spine and femurs were obtained by adding the signal intensity and infiltration or extent-of-involvement scores for these sites. A total BMB score of up to 8 could be given for the lumbar spine and a total BMB score of up to 8 could be given to the femurs. Thus, a total BMB score of up to 16 could be obtained by adding the lumbar and femoral BMB scores.
Statistics
The mean values for the total, femoral, and lumbar BMB scores with their
95% CIs were calculated for the baseline and final studies. The percentage
(with 95% CI) of patients whose BMB score improved by 2 points or more was
calculated. Standard equations were used for these calculations
[19]. CIs were not calculated
if the numbers were too low for this method to be valid.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Percentage of Patients Improving on Enzyme Replacement Therapy
In 39 subjects for whom baseline and final femoral BMB scores were
available, 24 subjects (62%, 95% CI, 47–76%) improved by 2 or more
points while on enzyme replacement therapy. Scores for 13 (50%; 95% CI,
31–69%) of the 26 for whom images obtained before and after enzyme
replacement therapy imaging were available improved by 2 or more points within
2 years of commencement of enzyme replacement therapy.
In 24 subjects for whom baseline and final lumbar BMB scores were available, the score in 16 subjects (67%; 95% CI, 48–85%) improved by 2 or more points.
Complete Data Set in 15 Patients
Table 3 shows the data for
15 subjects for whom the total BMB scores were available before enzyme
replacement therapy and at least one time point after the commencement of
enzyme replacement therapy. The total BMB scores improved by between 2 and 12
points (average, 6.3 points; median, 6 points). In six subjects, scores had
improved by greater than 2 points at 1 year after the commencement of enzyme
replacement therapy. CIs showed that there was a statistically significant
decrease in the average score at 2 years after commencement of enzyme
replacement therapy. In 12 cases, there was a decrease in score by 4 or more
points by 4 years after the commencement of enzyme replacement therapy. In one
case, there was a decrease in total BMB score of 10 by 3 years, but the score
increased again by 3 points at 4.75 years after the enzyme replacement therapy
dose was reduced from 34 to 29 U/kg per 2 weeks. In no case did a total BMB
score increase by more than 3 points on enzyme replacement therapy. The lumbar
BMB score increased by 2 points in one patient on enzyme replacement therapy
(subject 13 in Table 3), and
the femoral BMB increased by 2 points in two patients on enzyme replacement
therapy.
|
|
|
|
|
|
|
|
|
|
|
|
|
Two subjects developed vertebral compression fractures while on enzyme replacement therapy. In one of these patients, the fracture at T8 had occurred within 9 months of commencement of enzyme replacement therapy when the total BMB had improved by 2 points.
BMB Scores and Splenectomy
Although BMB scores were higher at baseline for patients who had undergone
splenectomy, of the subjects for whom data were available before commencement
of enzyme replacement therapy and for subjects at the final assessment after
enzyme replacement therapy, the CIs overlapped
(Table 4). Eight of 13
splenectomized subjects (62%, 95% CI, 36–87%) had extensive medullary
infarcts or avascular necrosis at baseline compared with 17 of 39 subjects
with a spleen in situ (44%, 95% CI, 29–59%).
|
BMB Scores and Infarction
The average baseline BMB score in the femurs for the 25 subjects with
extensive medullary infarcts or osteonecrosis was 6.5 (95% CI, 5.9–7.2)
and for those 24 subjects without infarcts or with small medullary infarcts
was 5.5 (95% CI, 4.8–6.1). The lumbar baseline BMB score for those 18
subjects with extensive infarcts was 6.2 (95% CI, 5.6–6.7) and for those
nine subjects without infarcts was 5.1 (95% CI, 4.1–6.1). The total BMB
score for those 18 subjects with infarcts was 12.9 (95% CI, 11.7–13.9),
and for those seven subjects without infarcts, it was 10.9 (95% CI,
8.8–12.9).
Higher baseline femoral and lumbar BMB scores were associated with a higher incidence of extensive medullary infarction or avascular necrosis. Twenty-seven of 36 subjects (75%, 95% CI, 61–89%) with a femoral BMB score of between 6 and 8 had extensive medullary infarction or avascular necrosis compared with three of 17 subjects (18%, 95% CI, 6–35%) with a femoral BMB score between 0 and 5. Eighteen of 22 subjects (82%, 95% CI, 66–97%) with a lumbar BMB score of between 6 and 8 had extensive medullary infarction or avascular necrosis compared with two of eight subjects (25%) with a lumbar BMB score between 0 and 5. The 95% CI was not given for the last proportion because the low numbers made the calculation invalid.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Numerous imaging methods are available to assess skeletal disease, and some have been reported in studies aimed at determining the extent of bone involvement in patients with Gaucher's disease. The various reported imaging approaches, including conventional radiography, nuclear medicine, CT, and bone densitometry, have all been shown to have limitations in quantifying the extent of marrow involvement in patients with Gaucher's disease.
In this study, we used a new scoring system to assess marrow infiltration in patients with Gaucher's disease that had been developed after correlating bone marrow changes with quantitative assessments made using QCSI [15]. We have shown that this scoring system can provide numeric data representing objectively measured improvement in bone marrow infiltration in Gaucher's disease. A significant improvement overall in mean scores was shown despite the fact that in 14 subjects, the first MR images were available 10–136 months (range, approximately 1–11 years; mean, > 4 years) after commencement of enzyme replacement therapy.
At 2 years after commencement of enzyme replacement therapy, for 15 subjects with pre- and posttreatment data sets, there was a statistically significant improvement in the mean total BMB score even allowing for the fact that not all had had repeat imaging at 2 years. In three subjects, a greater than 4 point decrease in score had occurred at or before 1 year after commencement of treatment.
In subject 13 of Table 3, the BMB score decreased initially from 14 to 4 at 3 years and was 7 at 4.75 years after commencement of enzyme replacement therapy and after a reduction in enzyme replacement therapy dose from 34 to 29 U/kg. It is difficult to know whether the difference between the last two scores is a definite change or it is due to an incorrect scoring of either the 3- or 4.75-year MRI examination. The femoral BMB score increased by 2 points in another two subjects; thus, until there is further validation of this scoring method, we recommend using a change in total BMB score of 4 or more as a significant indication of a change in the severity of skeletal involvement.
A scoring system has the advantage of giving a numeric value to guide decision-making. Ideally, all patients with Gaucher's disease would have a quantitative measure using QCSI, but because of its complexity, QCSI is not available at most sites. Other semiquantitative scoring systems have been reported [11, 13]. In the system reported by Terk et al. [8], the whole length of the femurs is evaluated, whereas in the system used by Poll et al. [13], the length of the leg from the hips to ankles was analyzed. In the so-called "Terk system" [8], a score of 0 was given for normal marrow, 1 for marrow that was of lower signal on T1- and T2-weighted images, and 2 for marrow signal that was decreased on T1-weighted images and increased on T2-weighted images. A score of 3 was given for a markedly heterogeneous marrow pattern. The system reported by Poll et al. [13] gave an "A" or "B" for morphology depending on whether the marrow there was of homogeneously decreased signal intensity (type A) or nonhomogeneous patchy decreased signal intensity (type B) on T1- and T2-weighted images. A score of 1–8 was then given for the extent of involvement depending on the number of anatomic areas involved. The advantages and disadvantages of these scoring systems and others have been extensively discussed by vom Dahl et al. [20] after a roundtable meeting of European experts in Gaucher's disease in September 2005.
The system reported by Maas et al. [16] requires evaluation of both the femurs and the lumbar spine to give a score out of 16 and analyze both signal intensity and extent of involvement. There are a number of potential advantages in the latter system: There is a greater range of signal intensity values evaluated, ranging from 1 to 5 for the lumbar spine and for the femurs based on comparison with QCSI scores. The score of 1–3 for extent of disease is based on the observed pattern of infiltration from milder forms of the disease to more severe infiltration and a higher score is given for more diffuse infiltration in the spine and also for diffuse involvement including infiltration of the fat around the basivertebral veins. In the femurs, a score of 1 is given for involvement of metaphyses and diaphyses only, a finding that occurs in milder disease, whereas involvement of the proximal and distal epiphyses results a higher score.
This scoring system allows separate evaluation of two individual bone marrow regions in the body. Extensive infarction is more common in the femurs than in the lumbar spine, so improvement in bone marrow infiltration in the femurs may be difficult to detect, whereas it might be seen in the lumbar spine. Further study is needed to determine whether patients with extensive infarction in the femurs or elsewhere should be treated differently compared with those without extensive infarction when there is improvement of marrow signal at another site. It is possible, however, that despite evidence of marrow improvement in the spine, bone disease remains severe; our experience with one subject suggested that any reduction of therapy in these subjects needs to be approached with caution.
Although it may be helpful to consider lumbar spine and femoral BMB data separately, evaluation of both will give a score out of 16, so a definite difference is more likely to be detected. In this study, a score of 6, 7, or 8 in either the lumbar spine or the femurs indicated severe disease using the outcome measure of extensive infarction. Lower scores of 1–3 may indicate residual Gaucher cell infiltration or may represent normal hematopoietic marrow.
We used fat-suppressed imaging for the femurs because it showed the bone marrow edema pattern more clearly. The BMB scoring system was originally described using non–fat-suppressed T2-weighted images [15]. It is inconclusive whether the categories for T2-weighted imaging were measuring exactly the same abnormalities, but from our experience we assumed that the categories of "small hyperintense foci" and of "extensive hyperintense foci," giving a score of 1 or 2 for T2-weighted imaging, corresponded to a similar severity of disease as did the BMB categories using non–fat-suppressed imaging.
Possible ways a BMB score may be used to adjust therapy include, first, using a higher dose of enzyme replacement therapy at commencement for those with a higher BMB score; second, dosage reduction for those with an improving BMB score; and, third, increasing the dose of enzyme replacement therapy in subjects in whom the BMB score does not decrease by 4 points. Further investigations will be needed to confirm these suggestions. This scoring system will also be useful in the evaluation of newer drugs for cases in which QCSI is not available. The BMB scoring method fulfilled the criteria for the preferred semiquantitative method for assessment of bone marrow infiltration at the recent roundtable meeting of European experts in Gaucher's disease [20].
Both reviewers in this study were experienced in the interpretation of MRI in patients with Gaucher's disease, having been responsible for assessing skeletal changes in a significant cohort of patients in nationally coordinated enzyme replacement therapy programs. It is likely that there is a learning curve in using this system and it may therefore be best applied at central review sites.
In clinical practice, we recommend using this scoring system in conjunction with descriptive imaging. In one case, in which involvement extended to the proximal femoral epiphyses before commencement of enzyme replacement therapy, the femoral BMB score increased by only 1 point. More extensive epiphyseal involvement is, however, of concern, and that finding would warrant an increase in or commencement of enzyme replacement therapy.
Limitations to this study include, first, that neither internor intraobserver agreement for the BMB score was evaluated, but this has been assessed in a previous study [15]. Second, MRI was performed at multiple sites, and the quality of the images varied, and third, a complete data set was available for only 15 subjects. Inclusion of other subjects was important, however, because the reviewers did not know which patients were on treatment at the time of the examination [4]. Because QCSI was not available for these patients, we were not able to compare against a quantitative scoring system. Despite these limitations, the study showed serial changes in the MRI marrow scores that appear to represent a measurable response to enzyme replacement therapy. Of particular interest was that significant changes in the BMB score were apparent within 2 years of commencement of enzyme replacement therapy.
In conclusion, we have shown the use of the newer Maas et al. [16] semiquantitative scoring method for the evaluation of bone marrow infiltration to assess the severity of skeletal involvement in patients with Gaucher's disease in the Australian adult Gaucher's disease population. This will facilitate the tailoring of dosage regimens in the future. Currently, a decrease in total BMB score of 4 points or of a femoral or lumbar BMB score of at least 2 points appears to indicate a definite improvement in skeletal disease and a measurable response to enzyme replacement therapy.
Acknowledgments
We are grateful to the treating physicians and radiologists for supplying
the patient data on which this review is based. The program is coordinated by
Kim Nguyen of the Life Saving Drugs Program.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
