Diagnostic Biomarkers of Parkinson's Disease
Diagnostic Biomarkers of Parkinson's Disease
Two articles in previous, recent issues of this journal, reported the impact of biomarkers on the differential diagnosis of early Parkinson's disease (PD) by using two complementary tools. Kupsch et al reported that DaTscan single photon emission computed tomographhy imaging resulted in a change of diagnosis of clinically uncertain parkinsonian syndromes at 4 weeks in 45% of the patients versus 9% of the controls who did not receive the DaTscan, resulting in an increase in confidence of the diagnosis. This study adds credence to similar conclusions from previous studies by using an excellent design with prospective follow-up and inclusion of randomised control subjects who did not receive the DaTscan. However, there are several caveats in using this diagnostic test in routine clinical practice. This and other studies examining the diagnostic validity of DaTscan used clinical diagnosis as the gold standard, usually with reassessment after several months to a few years by movement disorder experts. Therefore, the main benefit demonstrated by DaTscan is buying some time for an earlier diagnosis. The authors addressed the important question of whether an early change in diagnosis resulted in a change in clinical management. By the 12-week visit, 28% of patients in the DaTscan group started taking medications that were not originally planned. However, 22% of the controls also started taking medication, although only 12% had a change in diagnosis, suggesting that the medication was tried even without the benefit of revised diagnosis. The next logical question is whether the cost and potential risk of DaTscan with the subsequent change in clinical management are justified by improvement in the quality of life or optimisation of health resource use. The answer from this study is 'No'. The authors appropriately note that 1 year may not be sufficiently long to see these changes. In this regard, most clinicians would agree that the availability of neuroprotective agents would make an early diagnosis more compelling since it would make a stronger impact on the quality of life. Finally, one needs to bear in mind that few studies have examined the test-to-test reproducibility of DaTscan. Tolosa et al repeated the DaTscan in six cases that had a discrepancy between the DaTscan diagnosis and the clinical diagnosis. In four of these cases, the second reading confirmed the clinical diagnosis, which trumped the first DaTscan diagnosis.
DaTscan may facilitate an earlier diagnosis of parkinsonian syndromes, but transport imaging is expensive and does not reliably distinguish PD from other neurodegenerative parkinsonian syndromes. Are there biomarkers that could provide an accurate differential diagnosis, with less risk and at a low cost? Busse et al address the utility of a combined assessment of asymmetry of motor symptoms, hyposmia and hyperechogenecity of substantia nigra in improving specificity of PD diagnosis by differentiating it from other parkinsonism syndromes and essential tremors. The gold standard for the diagnosis in this paper was the British Brain Bank criteria for definite PD, which has 98% specificity when measured against autopsy findings. The specificity increased from 54%, arrived at by using asymmetry alone, to 84% with the addition of hyposmia and then to 94% with all three measures including substantia nigra hyperechogenecity. The diagnostic accuracy of these measures held up in a milder and younger validation cohort using clinical diagnosis made 1 year later as the gold standard. Similar to the findings of Kupsch et al, the main benefit of these measures was buying 1 year of time for an earlier diagnosis. Again, one needs to consider whether an early diagnosis of these conditions justifies the cost of these additional tests when treatment options are limited. Furthermore, if these tests exclude the possibility of PD, does it justify a possible missed opportunity for a therapeutic trial of levodopa? Most parkinsonian syndromes have at least a partial response to levodopa, at least in the early stages. Dopa-responsive dystonia is another example in which levodopa therapy should not be withheld. We eagerly await the availability of robust disease modifying therapies that could take advantage of an early diagnosis of PD and impact disease progression to justify the cost and effort of these additional tests.
Abstact and Introduction
Introduction
Two articles in previous, recent issues of this journal, reported the impact of biomarkers on the differential diagnosis of early Parkinson's disease (PD) by using two complementary tools. Kupsch et al reported that DaTscan single photon emission computed tomographhy imaging resulted in a change of diagnosis of clinically uncertain parkinsonian syndromes at 4 weeks in 45% of the patients versus 9% of the controls who did not receive the DaTscan, resulting in an increase in confidence of the diagnosis. This study adds credence to similar conclusions from previous studies by using an excellent design with prospective follow-up and inclusion of randomised control subjects who did not receive the DaTscan. However, there are several caveats in using this diagnostic test in routine clinical practice. This and other studies examining the diagnostic validity of DaTscan used clinical diagnosis as the gold standard, usually with reassessment after several months to a few years by movement disorder experts. Therefore, the main benefit demonstrated by DaTscan is buying some time for an earlier diagnosis. The authors addressed the important question of whether an early change in diagnosis resulted in a change in clinical management. By the 12-week visit, 28% of patients in the DaTscan group started taking medications that were not originally planned. However, 22% of the controls also started taking medication, although only 12% had a change in diagnosis, suggesting that the medication was tried even without the benefit of revised diagnosis. The next logical question is whether the cost and potential risk of DaTscan with the subsequent change in clinical management are justified by improvement in the quality of life or optimisation of health resource use. The answer from this study is 'No'. The authors appropriately note that 1 year may not be sufficiently long to see these changes. In this regard, most clinicians would agree that the availability of neuroprotective agents would make an early diagnosis more compelling since it would make a stronger impact on the quality of life. Finally, one needs to bear in mind that few studies have examined the test-to-test reproducibility of DaTscan. Tolosa et al repeated the DaTscan in six cases that had a discrepancy between the DaTscan diagnosis and the clinical diagnosis. In four of these cases, the second reading confirmed the clinical diagnosis, which trumped the first DaTscan diagnosis.
DaTscan may facilitate an earlier diagnosis of parkinsonian syndromes, but transport imaging is expensive and does not reliably distinguish PD from other neurodegenerative parkinsonian syndromes. Are there biomarkers that could provide an accurate differential diagnosis, with less risk and at a low cost? Busse et al address the utility of a combined assessment of asymmetry of motor symptoms, hyposmia and hyperechogenecity of substantia nigra in improving specificity of PD diagnosis by differentiating it from other parkinsonism syndromes and essential tremors. The gold standard for the diagnosis in this paper was the British Brain Bank criteria for definite PD, which has 98% specificity when measured against autopsy findings. The specificity increased from 54%, arrived at by using asymmetry alone, to 84% with the addition of hyposmia and then to 94% with all three measures including substantia nigra hyperechogenecity. The diagnostic accuracy of these measures held up in a milder and younger validation cohort using clinical diagnosis made 1 year later as the gold standard. Similar to the findings of Kupsch et al, the main benefit of these measures was buying 1 year of time for an earlier diagnosis. Again, one needs to consider whether an early diagnosis of these conditions justifies the cost of these additional tests when treatment options are limited. Furthermore, if these tests exclude the possibility of PD, does it justify a possible missed opportunity for a therapeutic trial of levodopa? Most parkinsonian syndromes have at least a partial response to levodopa, at least in the early stages. Dopa-responsive dystonia is another example in which levodopa therapy should not be withheld. We eagerly await the availability of robust disease modifying therapies that could take advantage of an early diagnosis of PD and impact disease progression to justify the cost and effort of these additional tests.
Source...