1
Journal of Alzheimer’s Disease xx (20xx) x–xx DOI 10.3233/JAD-132216 IOS Press
3
roo f
2
Cost-Effectiveness of the Use of Biomarkers in Cerebrospinal Fluid for Alzheimer’s Disease
or P
1
5
Cristina Valc´arcel Nazcoa,b,∗ , Lilisbeth Perestelo P´ereza,c , Jos´e Luis Molinuevod , Javier Marc,e , Iv´an Castillaa,f and Pedro Serrano Aguilara,c
6
a Evaluation
7
b Canary
4
12
Handling Associate Editor: Ramon Luengo-Fernandez
9 10
cte
Accepted 17 April 2014
dA
8
uth
11
Unit of the Canary Islands Health Service (SESCS), Tenerife, Spain Islands Foundation for Health and Research (FUNCIS), Tenerife, Spain c Health Services Research on Chronic Patients Network (REDISSEC), Bizkaia, Spain d Alzheimer’s disease and other cognitive disorders unit. IDIBAPS, Hospital Clinic, Barcelona, Spain e Clinical Management Service, Alto Deba Hospital, Mondragon, Spain f Spanish Consortium for Research on Epidemiology and Public Health (CIBERESP), Madrid, Spain
31
Keywords: Alzheimer’s disease, amyloid-, biomarkers, cost-effectiveness, diagnosis, phosphorylated tau, total tau
16 17 18 19 20 21 22 23 24 25 26 27 28 29
co
15
Un
14
rre
30
Abstract. Background: The use of cerebrospinal fluid (CSF) biomarkers could facilitate early detection of Alzheimer’s disease (AD) in patients with mild cognitive impairment (MCI) and the differential diagnosis between AD and non-AD dementias. The aim of this study is to assess the clinical and economic value of CSF biomarkers to diagnose AD. Objective: To determine the cost-effectiveness of the use of amyloid- peptide (A42 ), total tau, and phosphorylated tau proteins in CSF to diagnose AD in MCI and dementia patients. Methods: An economic evaluation was performed by means of cost-effectiveness analysis comparing two AD diagnostic alternatives: the combined determination of A42 proteins, total tau, and phosphorylated tau in CSF as biomarkers of AD, and the standard clinical diagnosis based on the Related Disorders Association (NINDS-ADRDA) criteria. A decision analytic model was developed to synthesize the identified evidence and to compare the costs and effectiveness associated with each diagnostic strategy. A probabilistic sensitivity analysis using 2nd order Monte Carlo simulations was performed. Subsequently, acceptability curves were calculated and ANCOVA models were applied to the results of the Monte Carlo simulations in order to identify the parameters that led greater variability in the model outcomes. Results: The use of CSF biomarkers as an early diagnostic strategy of AD in MCI patients is a dominant alternative (less costly and more effective strategy than diagnostic criteria). In dementia patients, although there is a higher uncertainty, biomarkers in CSF seem a more cost-effective alternative than diagnostic criteria. Conclusions: Detecting AD in MCI patients by determining A42 , total tau, and phosphorylated tau proteins biomarkers in CSF is a cost-effective diagnostic alternative. No conclusive results were obtained on dementia patients.
13
∗ Correspondence
to: Cristina Valc´arcel Nazco, Servicio de Evaluaci´on del Servicio Canario de la Salud (SESCS), Camino Candelaria, 44, 38109 El Rosario, Santa Cruz de Tenerife, Canary Islands, Spain. Tel.: +34 922 68 40 19; Email: cristina.valcar [email protected].
ISSN 1387-2877/14/$27.50 © 2014 – IOS Press and the authors. All rights reserved
39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82
roo f
38
or P
37
in research; however, they have now fallen behind the unprecedented growth of scientific knowledge. For this reason, in 2007, the International Working Group (IWG) new research criteria [15] introduced a new conceptualization of AD and created a framework to establish a specific diagnosis in the pre-dementia or prodromal stage of the disease [16]. There is increasing consensus to understand AD as a clinical-biological entity, in which biomarkers, especially pathophysiological markers revealing the underlying pathology, represent the biological counterpart of the diagnosis, and specific symptoms, such as episodic memory deficits, account for the clinical one. In 2011, the National Institute of Aging Alzheimer Association (NIA-AA) workgroups also published biomarker supported diagnostic criteria to cover all disease stages [17]. These new criteria, which allow different level of diagnostic certainty based on biomarker results, permit detecting the asymptomatic or preclinical stage [18], the predementia or prodromal stage, termed MCI due to AD [19], and the dementia stage, named dementia due to AD [7]. However, the diagnosis in the predementia stage, applying the IWG criteria [15] or the MCI due to AD criteria [19], are considered to be applicable only in the research setting, since although some research supports the notion that biomarkers may help in predicting conversion to dementia [20, 21], there is yet no clear conviction that the time of AD diagnosis can be reliably advanced [22]. The appropriate selection, detection and visualization of biomarkers can aid in the diagnosis of AD and may prevent a significant number of false positive diagnoses by clinicians using clinical guidelines alone. Although significant advances have been made in the field of neuroimaging, biomarkers based on CSF are at present the most convenient for studying disease progression [11, 23, 24]. Among the better studied biomarkers are the quantification of the amyloid- peptide (A42 ) and tau protein (total tau and phosphorylated tau proteins) in the CSF. Accumulation of amyloid plaques and neurofibrillary tangles probably starts 20 to 30 years before the clinical onset of the disease. Therefore, CSF biomarkers are the most suitable candidates to facilitate AD diagnosis in the very early stages of the disease, long before symptoms onset. Moreover, since it may be optimal to treat the neuropathology as early as possible, biomarkers of preclinical AD are likely to play a pivotal role in the development of the next generation of therapies [24]. The diagnostic sensitivity and specificity are between 80–90% for each biomarker separately [25]. Combining the results of all three biomarkers is now
dA
36
cte
35
rre
34
Alzheimer’s disease (AD) is the leading cause of dementia in the elderly [1] involving up to 70% of dementia cases. The reported prevalence in Europe is 6.2% for people over 60 years old and it is estimated a 40% increase in number of people with AD by 2030 [2]. The health and social relevance of AD and the need of effective treatments have generated an increased interest on the development and evaluation of valid diagnostic procedures for its early detection. To date, AD is diagnosed at the dementia stage of the disease when cognitive impairment interferes with the patient’s functional capacities for daily living [3]. This clinical threshold does vary among individuals, depending on factors such as premorbid cognitive performance, level of intelligence, complexity of everyday activities, or level of perception of the informant. Currently, the standard diagnostic procedure consists of a set of internationally validated clinical criteria. The most frequently used criteria are those defined in the 4th Diagnostic and Statistical Manual of Mental Disorders (DSM-IV-TR) [4, 5], the 10th International Classification of Disease (ICD-10) [6], and in the guidelines of the National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Association (NINDS-ADRDA guidelines) [3, 7]. These criteria are based on the detection of the dementia syndrome and the classic features of AD, being more oriented to the exclusion of other non-degenerative causes of dementia. Although diagnosis based on these criteria has proven to be very sensitive to AD when compared to other dementias, its specificity reaches only 50 to 60% [8]. Hence, using only clinical criteria to diagnose AD may lead to late, and unnecessary and expensive treatment due to this substantial number of false positives. An earlier and more specific diagnosis may be of help to patients and clinicians, regardless that the underlying biological process cannot be stopped [9], by contributing to guiding therapy and properly advising patients and families. These reasons explain the research interest in developing valid neuropsychological, neuroimaging, blood, or cerebrospinal fluid (CSF) markers that could allow both the early detection of AD in patients with mild cognitive impairment (MCI) [10–13] and the differential diagnosis between AD and non-AD dementias [10, 12, 14]. The NINCDS–ADRDA and the DSM-IV-TR criteria for AD are the prevailing diagnostic standards
co
33
INTRODUCTION
Un
32
C.V. Nazco et al. / Cost-Effectiveness of the Use of Biomarkers in Cerebrospinal Fluid
uth
2
83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134
C.V. Nazco et al. / Cost-Effectiveness of the Use of Biomarkers in Cerebrospinal Fluid
140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159
160
161
METHODS Literature search
172
Model overview
165 166 167 168 169 170
173 174 175 176 177 178 179 180 181
co
164
Un
163
rre
171
A literature search was performed, without temporal restrictions, to identify systematic reviews, meta-analysis, and economic evaluations on the diagnostic performance of CSF biomarkers of AD, in the databases MEDLINE and PREMEDLINE, Centre for Reviews and Dissemination and EMBASE. The following search terms were used: (“alzheimer’s disease diagnosis” or “alzheimer’s disease”) and (“abeta-42” or “T-tau” or “P-tau” or “tau” or “phospho-tau” or “phosphorylated tau”).
162
roo f
139
or P
138
Figure 1 shows a schematic outline of the modeling approach. The model assesses costs from the perspective of the Spanish National Health System (NHS) expressed in Euros (D ) at a 2013 base price. The principal effectiveness measure used was “appropriate diagnosis” that would enable correct information to apply symptomatic treatment with donepezil. Several studies have shown that acetylcholinesterase inhibitors like donepezil may reduce costs and even delay institutionalization but do not have any impact on patient survival [26]. Therefore, no QALYs were calculated because donepezil does not modify the natural history of AD. Consequently, the health outcomes of the two alternatives were considered to be the same. It should be noted that, in the diagnostic criteria alternative, many patients with dementias not associated with AD are treated for AD and do not obtain any benefit. Several trials have revealed the benefits of acetylcholinesterase inhibitors, such as donepezil on cognition, activities of daily living, and global functioning. Additionally, the cost-effectiveness of this treatment has been proved in economic studies [27]. Results were obtained separately for two clinical scenarios. The first scenario (Scenario 1) considered a hypothetic cohort of patients aged 60 and older with MCI; the second scenario (Scenario 2) assumed a hypothetic cohort of patients aged 60 and older, with symptoms of dementia. A lifetime horizon was used in the analysis, and costs were discounted at 3%. The model was implemented in TreeAge Pro 2009 Healthcare (TreeAge Software, Inc., USA).
uth
137
considered by some authors as the best biomarker of AD [24]. Though the diagnostic sensitivity of its combined determination is very high, its specificity is limited, as they might be also altered in other neurodegenerative diseases [14]. There are several misunderstandings in how biomarkers should be interpreted from a clinical perspective, in how the new research criteria could be applied in clinical practice, and in the ethical background supporting early diagnosis. This existing uncertainty demands an urgent need for health technology assessment to address the evaluation of the diagnostic strategies including biomarkers for AD compared to current clinical practice. Therefore, the aim of this study is to assess the clinical and economic value of CSF biomarkers for the early diagnosis and detection of AD; being the main objective to determine the cost-effectiveness of the use of A42 , total tau, and phosphorylated tau proteins in CSF for early detection of AD and confirmation of the diagnosis. The objective of our study was to determine the cost-effectiveness of the use of A42 , total tau, and phosphorylated tau proteins in CSF to diagnose AD in MCI and dementia patients.
dA
136
cte
135
3
An economic evaluation was performed by means of cost-effectiveness analysis comparing two diagnostic alternatives for AD: the combined determination of A42 , total tau, and phosphorylated tau proteins in CSF as biomarkers of AD and the standard clinical diagnosis based on the NINDS-ADRDA criteria [3]. A decision analytic model was developed to synthesize the evidence identified and to compare the costs and effectiveness associated with each technology.
Model inputs Parameters Model inputs were estimated from the literature [11, 12, 29, 30]. Table 1 shows the values and information sources of the most important parameters used in the base case for each scenario. In the first scenario, the cost-effectiveness analysis comprehends that, by definition, MCI patients cannot be diagnosed of AD at the time of the first evaluation and analysis; therefore the probability of detecting a suspected case (see below) in this scenario implicitly implies the probability of a given patient to develop AD in the future. To estimate the probability of detecting a suspected case of AD in each arm of the model, we proceeded as follows: being P AD the “Probability of developing AD”, P suspected AD the “Probability of detecting
182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215
216
217 218 219 220 221 222 223 224 225 226 227 228 229 230 231
C.V. Nazco et al. / Cost-Effectiveness of the Use of Biomarkers in Cerebrospinal Fluid
or P
roo f
4
uth
Fig. 1. Decision tree model structure. P suspected AD biomarkers = Probability of suspected case of AD (biomarkers). P suspected AD diagnostic criteria = Probability of suspected case of AD (diagnostic criteria). PPV biomarkers = Positive predictive value of biomarker strategy. NPV biomarkers = Negative predictive value of biomarker strategy. PPV diagnostic criteria = Positive predictive value of diagnostic criteria. NPV diagnostic criteria = Negative predictive value of diagnostic criteria. #=complementary probability.
Table 1 Base case estimates
Item
Un
co
rre
cte
SCENARIO 1 (MCI patients) AD incidence AD prevalence Sensitivity of biomarkers Specificity of biomarkers Sensitivity of diagnostic criteria Specificity of diagnostic criteria Probability of suspected case of AD (biomarkers) Probability of suspected case of AD (diagnostic criteria) SCENARIO 2 (Dementia patients) AD incidence AD prevalence Sensitivity of biomarkers Specificity of biomarkers Sensitivity of diagnostic criteria Specificity of diagnostic criteria Probability of suspected case of AD (biomarkers) Probability of suspected case of AD (diagnostic criteria)
dA
Item
BIOMARKERS First neurology outpatient appointment Lumbar puncture Reactive per sample (ELISA kits) Hour worked of laboratory technician Hour worked of physician Subsequent neurology outpatient appointment Treatment with donepezil during lifetime DIAGNOSTIC CRITERIA First neurology outpatient appointment Subsequent neurology outpatient appointment Treatment with donepezil during lifetime
Estimate
Source
0.0319 0.086 0.81 0.87 0.87 0.58 0.1517 0.4344
[29] [29] [11] [11] [30] [30] [11, 29] [29, 30]
0.0408 0.1226 0.84 0.71 0.873 0.443 0.3124 0.5699
[30] [30] [12] [12] [30] [30] [12, 29] [29, 30]
Table 2 Resource use and unit costs (2013 D ) Resource use
Unit cost
Source
1 1 3 1 1 1 10 mg daily
D 170.82 D 158.28 D 22.08 D 7.99 D 16.55 D 102.49 D 2.33 (annual cost of D 850.06)
[32] [32] ELISA kits distributor CHS CHS [32] [33]
1 1 10 mg daily
D 170.82 D 102.49 D 2.33 (annual cost of D 850.06)
[32] [32] [33]
ELISA, Enzyme-linked immunosorbent assay; CHS, Canarian Health Service.
C.V. Nazco et al. / Cost-Effectiveness of the Use of Biomarkers in Cerebrospinal Fluid
236
237
238
P AD = P suspected AD∗ P TP ∗
+(1 − P suspected AD) (1 − P TN)
239
244 245 246
247
248
249
250
251 252 253
254 255 256 257 258 259 260
261 262 263 264 265 266 267 268 269 270 271 272 273 274 275
PPV = Sens ∗ Prev/Sens ∗ Prev + (1 − Esp) ∗ (1 − Prev) NPV = Esp ∗ (1 − Prev)/Esp ∗ (1 − Prev) + (1 − Sens) ∗ Prev where Sens = Sensibility of the diagnostic strategy, Esp = Specificity of the diagnostic strategy, and Prev = Prevalence of AD.
A sensitivity analysis was conducted in order to determine whether the results of the study changed when the values of certain variables were altered. The difference in cost between strategies was divided by the difference in effects to measure the cost per case detected, expressed as the incremental costeffectiveness ratio (ICER). In order to account for the uncertainty in key model parameters, deterministic and probabilistic sensitivity analysis using 10,000 Monte Carlo simulations [34] were performed. One suggested method to represent this uncertainty is costeffectiveness acceptability curves [35]. Acceptability curves represent the probability that a given intervention is cost-effective at different values of willingness to pay (WTP) per effectiveness unit. Table 3 shows the parameters used in the sensitivity analysis [11, 12, 24, 29, 30, 36, 37]. The cost-effectiveness of an alternative depends on the WTP of the payer for each additional effectiveness unit. Since no references have been found for the Spanish NHS WTP, results have been presented for WTP of 10,000 and 1,000 D for correctly diagnosed case. Finally, the percentage variability of the costs and effects explained by each parameter of the model was calculated using ANCOVA models. The advantage of this method compared to a non-probabilistic sensitivity analysis is that, for each parameter set, variations in the other model parameters are taken into account.
cte
243
Clinical effectiveness Effectiveness was measured in natural clinical units in terms of correctly diagnosed cases of AD. Therefore, the measure of effectiveness in each arm of the model is directly related to the sensitivity and specificity of each diagnostic strategy, that is, true positives and true negatives of each alternative compared.
rre
242
co
241
Thus, eliminating the unknown P suspected AD, we obtained the probability of detecting a suspected case in each diagnostic strategy based on sensitivity and specificity. The positive and negative predictive values (PPV and NPV) of each diagnostic strategy were estimated, respectively, by the formulae:
Resource use and unit costs According to the perspective adopted for the analysis, only direct health care costs such as drugs (donepezil), medical visits and diagnostic procedures, were considered. As mentioned before, treatment with donepezil was assumed to have no influence on patient survival, so estimates of time to death in the model were assumed to be identical for treated and untreated patients. Median survival with AD was estimated at 7.1 years (95% CI 6.7–7.5 years) [31] according to the literature. Treatment with donepezil continues right up until death. Table 2 shows resource use per patient and unit costs used in the analysis [32, 33].
Un
240
276
roo f
235
Sensitivity analysis
or P
234
a suspected case of AD”, P TP the “Probability of obtaining a True Positive (sensitivity of diagnostic strategy)”, and P TN the “Probability of obtaining a True Negative (specificity of diagnostic strategy)”, then
uth
233
dA
232
5
277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305
RESULTS
306
Base case results
307
Scenario 1 (patients with MCI) shows a lower average cost per patient for biomarkers use than for conventional diagnostic clinical criteria (1,336.06 D versus 3,167.39 D , respectively). This is due to the fact that conventional clinical criteria misdiagnosed more cases of AD treated with donepezil. In addition, biomarkers in CSF have a likelihood of accurate detection of 88,73%, which means that most MCI patients with a positive biomarker profile will develop AD in the future, whereas this probability falls below 52% for the alternative strategy. Base case results for MCI patients reveal that CSF biomarkers for early AD detection are a dominant diagnostic strategy. The average cost of a dementia patient (Scenario 2) diagnosed by CSF biomarkers is higher than the average cost of a dementia patient diagnosed
308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323
6
C.V. Nazco et al. / Cost-Effectiveness of the Use of Biomarkers in Cerebrospinal Fluid Table 3 Parameters used in sensitivity analysis Distribution (Mean ± Error)
SCENARIO 1 (MCI patients) AD incidence AD prevalence Sensitivity of biomarkers Specificity of biomarkers Sensitivity of diagnostic criteria Specificity of diagnostic criteria SCENARIO 2 (Dementia patients) AD incidence AD prevalence Sensitivity of biomarkers Specificity of biomarkers Sensitivity of diagnostic criteria Specificity of diagnostic criteria BOTH SCENARIOS Cost of first neurology outpatient appointment Cost of subsequent neurology outpatient appointment Cost, per hour, of a physician Cost, per hour, of a laboratory technician Cost of lumbar puncture Discount Annual cost of donepezil
BETA (0.0319 ± 0.02) BETA (0.086 ± 0.02) BETA (0.81 ± 0.081) BETA (0.87 ± 0.043) BETA (0.859 ± 0.085) BETA (0.471 ± 0.047)
or P
BETA (0.0408 ± 0.028) BETA (0.1226 ± 0.03) BETA (0.84 ± 0.071) BETA (0.71 ± 0.106) BETA (0.873 ± 0.087) BETA (0.443 ± 0.04)
329
cte
328
rre
327
[29, 30] [29] [12] [12] [30] [30] [32, 37] [32, 37] CHS CHS [32] 0%–3% D 680.048–D 850.06
Results of sensitivity analysis
330
The results of the deterministic sensitivity analysis are shown in Table 4. In all cases, the use of CSF biomarkers resulted a dominant strategy, i.e., less costly and more effective than the alternative strategy.
Table 4 Results of deterministic sensitivity analysis
Parameter SCENARIO 1 (MCI patients) Discount Annual cost of donepezil SCENARIO 2 (Dementia patients) Discount Annual cost of donepezil
co
326
by conventional clinical criteria (2,198.33 D versus 3,330.28 D , respectively). In this scenario, biomarkers in CSF have a likelihood of accurate detection of 75.66%; whereas the probability of detecting a suspected case of AD in the alternative strategy is 52%.
Parameter SCENARIO 1 (MCI patients) Cost Accurate diagnosis∗ SCENARIO 2 (Dementia patients) Cost Accurate diagnosis∗ ∗ Probability
Un
325
[29] [29] [11] [11] [30] [30]
dA
uth
GAMMA (170.82 ± 42.705) GAMMA (102.49 ± 25.62) GAMMA (16.55 ± 4.965) GAMMA (7.99 ± 2.397) GAMMA (158.28 ± 47.48)
CHS, Canarian Health Service.
324
Source
roo f
Item
Base case
Value
ICER
3% D 850.06
0% D 680.048
Dominant Dominant
3% D 850.06
0% D 680.048
Dominant Dominant
Table 5 Results of probabilistic sensitivity analysis
Biomarkers
Diagnostic criteria
Incremental
ICER
1,334.30 D [906.79; 1,871.62] 0.8857 [0.7965; 0.9506]
3,166.96 D [2,672.22; 3,649.06] 0.5184 [0.4234; 0.6144]
D −1,832.65 [−2,473.48; −1,142.65] 0.3672 [0.2418; 0.4833]
Dominant
2,198.05 D [1,256.53; 3,346.87] 0.7541 [0.5487; 0.9119]
3,331.87 D [2,854.64; 3,790.06] 0.5149 [0.4169; 0.6170]
−1,133.82 D [−2,176.82; 77,95] 0.2392 [0.0216; 0.4215]
Dominant [Dominant; 3,607.54]
of achieving the right diagnosis.
331 332 333 334
7
Un
co
rre
cte
dA
uth
or P
roo f
C.V. Nazco et al. / Cost-Effectiveness of the Use of Biomarkers in Cerebrospinal Fluid
Fig. 2. Incremental cost-effectiveness plane.
335 336 337 338 339 340
Table 5 presents the results of the probabilistic sensitivity analysis for the two alternative scenarios. The results from scenario 1 suggest similar conclusions to the base case analysis. The use of CSF biomarkers as an early diagnostic strategy of AD in patients with MCI is a dominant alternative. In scenario 2, the average ICER
value highlights the uncertainty over the decision to take. Figure 2 shows the incremental cost-effectiveness plane for each considered scenario. These graphs represent the cost and effectiveness pairs obtained from each simulation model, and the average ICER. As seen
341 342 343 344 345 346
C.V. Nazco et al. / Cost-Effectiveness of the Use of Biomarkers in Cerebrospinal Fluid
uth
or P
roo f
8
Fig. 3. Acceptability curve. scenario 2: Dementia patients.
370
DISCUSSION
352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368
371 372 373 374 375 376
cte
351
rre
350
co
349
Un
348
cial interventions [43]. However, we have not been able to identify studies on the cost-effectiveness of diagnostic procedures focused on early detection of AD when only signs of MCI are visible. This study has developed an economic evaluation that compares, in terms of diagnostic efficiency, two alternatives for the detection of AD. We have considered two scenarios with different patients: patients with MCI who have no symptoms of dementia, and patients with those symptoms. Although there are not available effective drugs to change AD natural history, symptomatic drug treatment is widely used according to clinical criteria. The use of biomarkers in CSF for early and more valid diagnosis of AD plays a leading role as it contributes to an accurate and more efficient diagnosis by means of objective and reliable procedures. These findings could be of interest for clinical research, to inform patients and families, and to guide earlier treatment decisions when more effective alternatives exist. This economic evaluation does not analyze the efficiency of diagnosis of MCI leading to AD or the early diagnosis of AD in terms of health-related quality of life (HRQOL). Instead, this economic evaluation is interested in the diagnostic efficiency of biomarkers for the detection of AD, using “cases detected and treated correctly” as the measure of effectiveness. Although using “cases detected and treated correctly” has the limitation of being an intermediate outcome, it was selected because the objective of this report was to assess the diagnostic efficiency of protein biomarkers in CSF.
dA
369
in Fig. 2, the use of biomarkers in CSF is less costly and more effective than diagnostic criteria in MCI patients. For dementia patients, a confidence interval for ICER of [Dominant; 3,607.54] was obtained. The acceptability curve of Scenario 2 for different WTP values is presented in Fig. 3. If we draw a vertical line on the graph at a given WTP value, the curve represents the probability that the use of biomarkers in CSF is a cost-effective alternative. Therefore, in this scenario (dementia patients), the probability of accepting biomarkers as a cost-effective alternative is approximately 80% for 1,000 D WTP. The acceptability curve for scenario 1 is not presented because the biomarker strategy is a dominant alternative. The ANCOVA analysis results are shown in Fig. 4. In scenario 1, the parameter that has the greatest impact on both costs and effectiveness is the specificity of the conventional clinical diagnostic criteria. Specificity of biomarker strategy also has a remarkable impact on both the cost and effectiveness variability. When focusing on scenario 2, the parameter that has a greater impact on both costs and effects is the specificity of biomarkers in CSF.
347
According to our results, the use of A42 , total tau, and phosphorylated tau proteins tests in CSF is clearly cost-effective in MCI patients but is not conclusive in dementia patients. Previous economics evaluations reported on the cost-effectiveness of AD treatments, including both pharmacologic [38–42] and psychoso-
377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408
9
Un
co
rre
cte
dA
uth
or P
roo f
C.V. Nazco et al. / Cost-Effectiveness of the Use of Biomarkers in Cerebrospinal Fluid
409 410 411 412 413 414 415 416 417
Fig. 4. Results of ANCOVA analysis.
To the best of our knowledge, this is the first study to analyze the diagnostic efficiency of protein biomarkers in CSF as compared to routine clinical diagnosis of AD. Previously, a study published a protocol designed to conduct an economic evaluation of these biomarkers in CSF to improve HRQOL in the hypothetic context of effective drugs being developed [44]. The results of this study will be published in two years and they will be theoretic.
The results of this economic evaluation prove the potential cost-effectiveness of the combined use of AD CSF biomarkers against the application of standard clinical criteria in patients with MCI who will develop AD later. For this group of patients, a dominant ICER was obtained. According to these results, the use of biomarkers in CSF is a cost-effective diagnostic alternative. The very low error probabilities given by the acceptability curves in patients with MCI, makes
418 419 420 421 422 423 424 425 426
C.V. Nazco et al. / Cost-Effectiveness of the Use of Biomarkers in Cerebrospinal Fluid
434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473
476 477
cte
433
rre
432
co
431
Un
430
roo f
Limitations
429
and sensitivity of the standard clinical diagnostic criteria for accurate diagnosis of AD. The uncertainty surrounding these parameter estimates influences the inability to determine which diagnostic strategy is more efficient for dementia patients. Another limitation to consider relates to the measure of effectiveness selected. In general, the outcome “cases detected and treated correctly” is considered an intermediate step in economic analysis. As mentioned above, no QALYs were calculated because donepezil does not modify the natural history of AD given that several studies have shown that acetylcholinesterase inhibitors do not have any impact on patient survival [26]. Moreover, this measure responds to the primary purpose of this economic evaluation of studying the diagnostic efficiency of protein biomarkers in CSF. Therefore, the model do not take into account variations in disease progression, as this would be the same in both strategies, but it does consider the cost savings related with avoided unnecessary drug treatment to patients wrongly diagnosed. The exclusion of other costs related to institutionalization delay could be considered a limitation. However, as biomarkers diagnostic strategy is dominant in MCI patients, the addition of these costs would reaffirm the conclusions of our study. Last but not least, the absence of a threshold value for the WTP in the Spanish context adds extra uncertainty to the conclusions of this study. This limitation is overcome in this study by presenting the results for different values of hypothetic WTP thresholds. According to the WTP taken as a reference, the results may favor or go against the efficiency of protein biomarkers in CSF as a diagnostic strategy for patients with symptoms of dementia.
or P
475
428
dA
474
CSF biomarkers a cost-effective procedure compared to conventional clinical diagnostic criteria. Variations in the specificity of clinical diagnostic criteria and in the specificity of biomarkers may, however, alter these results because these parameters have a large impact on model results. However, neither the prevalence of AD (nor incidence) nor biomarkers sensitivity affect the results of this model. The results obtained in MCI patients entail significant challenges for research on the diagnostic and therapeutic process of the early stages of AD. Although exact figures on the proportion of people with MCI who may progress to dementia are not known, there is consensus on the fact that MCI patients have a higher incidence of dementia than the general population [19, 45]. For patients with symptoms of dementia, the uncertainty around the parameters of the economic model developed is high. For this group of patients with suspected AD, the results obtained do not enable us to assure that biomarkers in CSF are more cost-effective than standard diagnostic criteria in terms of diagnostic efficiency. The parameters with more influence on the economic evaluation for dementia patients are biomarkers specificity and the cost of lumbar puncture. Precisely, the cost of lumbar puncture is highly variable among Spanish regions. This is an important element to be taken into account should standardization of this diagnostic technique on a national scale become an objective. The results of this economic evaluation may help initiatives such as The Alzheimer’s Disease Neuroimaging Initiative (ADNI), which attempts to identify and integrate information on diagnostic procedures based on neuroimaging and biomarkers [46]. Our results may also be useful for studies such as the Development of Screening Guidelines and Clinical Criteria for Predementia AD (DESCRIPA), which aims to develop guidelines for screening AD pre-dementia states in the general population [47]. Therefore, this new technology could be cost-effective in MCI patients from the perspective of the validity and efficiency of new diagnostic procedures to facilitate clinical and evaluative research on the effectiveness of new therapeutic interventions, in addition to informing patients and families and thereby assisting their earlier overall arrangements and, hopefully, a shared decision making process.
427
uth
10
The main limitation of this economic evaluation was the difficulty to find quality estimates of the specificity
CONCLUSIONS According to the results of this economic evaluation, detecting AD in patients with MCI by the determination of A42 proteins, total tau, and phosphorylated tau in CSF as biomarkers is a cost-effective alternative from the perspective of the Spanish NHS. However, since there is no defined WTP in Spain for this unit of effectiveness, these results should be treated with caution. When applying the same technique to patients with symptoms of dementia, there is a higher uncertainty in the results. Nevertheless, biomarkers in CSF still seem a more cost-effective alternative than diagnostic criteria.
478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512
513
514 515 516 517 518 519 520 521 522 523 524 525 526
C.V. Nazco et al. / Cost-Effectiveness of the Use of Biomarkers in Cerebrospinal Fluid
530 531 532 533 534 535 536
[14]
[15]
[16]
540
[2]
541 542 543
[3]
544 545 546 547 548
[4]
549 550 551
[5]
552 553
[6]
554 555 556
[7]
558 559 560 561 562 563 564
[8]
565 566 567 568 569
[9]
570 571 572 573
[10]
574 575 576 577
[11]
578 579 580 581 582 583 584
[18]
[19]
rre
557
[17]
dA
539
Khachaturian ZS (1985) Diagnosis of Alzheimer’s disease. Arch Neurol 42, 1097-1105. Alzheimer Disease, International (2010) World Alzheimer Report 2010: The Global Economic Impact of Dementia, Alzheimer Disease International, London. McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM (1984) Clinical diagnosis of Alzheimer’s disease: Report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology 34, 939-944. American Psychiatric Association (2000) Diagnostic and Statistical Manual of Mental Disorders, 4th edition. American Psychiatric Association, Washington DC. Lopez Ibor J (2005) DSM-IV-TR. Manual Diagnostico y Estadistico de los Trastornos Mentales, Masson, Barcelona. World Health, Organization (1992) The ICD-10 Classification of Mental and Behavioural Disorders, World Health Organization, Geneva. McKhann GM, Knopman DS, Chertkow H, Hyman BT, Jack CR, Kawas CH, Klunk WE, Koroshetz WJ, Manly JJ, Mayeux R, Mohs RC, Morris JC, Rossor MN, Scheltens P, Carrillo MC, Thies B, Weintraub S, Phelps CH (2011) The diagnosis of dementia due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7, 263-269. Lim A, Tsuang D, Kukull W, Nochlin D, Leverenz J, McCormick W, Bowen J, Teri L, Thompson J, Peskind ER, Raskind M, Larson EB (1999) Clinico-neuropathological correlation of Alzheimer’s disease in a community-based case series. J Am Geriatr Soc 47, 564-569. Molinuevo JL, Berthier ML, Rami L (2011) Donepezil provides greater benefits in mild compared to moderate Alzheimer’s disease: Implications for early diagnosis and treatment. Arch Gerontol Geriatr 52, 18-22. Mitchell AJ (2009) CSF phosphorylated tau in the diagnosis and prognosis of mild cognitive impairment and Alzheimer’s disease: A meta-analysis of 51 studies. J Neurol Neurosurg Psychiatry 80, 966-975. Monge-Argil´es JA, S´anchez-Pay´a J, Mu˜noz-Ruiz C, Pampliega-P´erez A, Montoya-Guti´errez J, Leiva-Santana C (2010) Biomarcadores en el l´ıquido cefalorraqu´ıdeo de pacientes con deterioro cognitivo leve: Metaan´alisis de su capacidad predictiva para el diagn´ostico de la enfermedad de Alzheimer. Rev Neurol 50, 193-200. Bloudek LM, Spackman DE, Blankenburg M, Sullivan SD (2011) Review and meta-analysis of biomarkers and diag-
cte
[1]
[12]
co
538
REFERENCES
Un
537
roo f
529
The authors gratefully acknowledge the contribution of Cristobal Carnero and Norberto Rodr´ıguez in the design of the economic evaluation model. The authors also thank Laura Vallejo Torres for the review of the manuscript. This study was funded by Spanish Health Ministry (Ministerio de Sanidad, Servicios Sociales e Igualdad). Authors’ disclosures available online (http://www.jalz.com/disclosures/view.php?id=2284).
[13]
nostic imaging in Alzheimer’s disease. J Alzheimers Dis 26, 627-645. Diniz BS, Pinto J´unior JA, Forlenza OV (2008) Do CSF total tau, phosphorylated tau, and beta-amyloid 42 help to predict progression of mild cognitive impairment to Alzheimer’s disease? A systematic review and meta-analysis of the literature. World J Biol Psychiatry 9, 172-182. Van Harten AC, Kester MI, Visser PJ, Blankenstein MA, Pijnenburg YA, van der Flier WM, Scheltens P (2011) Tau and p-tau as CSF biomarkers in dementia: A meta-analysis. Clin Chem Lab Med 49, 353-366. Cummings JL, Dubois B, Molinuevo JL, Scheltens P (2013) International work group criteria for the diagnosis of Alzheimer disease. Med Clin North Am 97, 363-368. Dubois B, Feldman HH, Jacova C, Dekosky ST, BarbergerGateau P, Cummings J, Delacourte A, Galasko D, Gauthier S, Jicha G, Meguro K, O’brien J, Pasquier F, Robert P, Rossor M, Salloway S, Stern Y, Visser PJ, Scheltens P (2007) Research criteria for the diagnosis of Alzheimer’s disease: Revising the NINCDS-ADRDA criteria. Lancet Neurol 6, 734-746. Jack CR, Albert MS, Knopman DS, McKhann GM, Sperling RA, Carrillo MC, Thies B, Phelps CH (2011) Introduction to the recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7, 257-262. Sperling RA, Aisen PS, Beckett LA, Bennett DA, Craft S, Fagan AM, Iwatsubo T, Jack CR Jr, Kaye J, Montine TJ, Park DC, Reiman EM, Rowe CC, Siemers E, Stern Y, Yaffe K, Carrillo MC, Thies B, Morrison-Bogorad M, Wagster MV, Phelps CH (2011) Toward defining the preclinical stages of Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7, 280-292. Albert MS, DeKosky ST, Dickson D, Dubois B, Feldman HH, Fox NC, Gamst A, Holtzman DM, Jagust WJ, Petersen RC, Snyder PJ, Carrillo MC, Thies B, Phelps CH (2011) The diagnosis of mild cognitive impairment due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7, 270-279. Ballard C, Gauthier S, Corbett A, Brayne C, Aarsland D, Jones E (2011) Alzheimer’s disease. Lancet 377, 1019-1031. Buchhave P, Minthon L, Zetterberg H, Wallin AK, Blennow K, Hansson O (2012) Cerebrospinal fluid levels of -amyloid 1-42, but not of tau, are fully changed already 5 to 10 years before the onset of Alzheimer dementia. Arch Gen Psychiatry 69, 98-106. Prince M, Bryce R, Ferri C (2011) World Alzheimer Report 2011: The Benefits of Early Diagnosis and Intervention. Alzheimer’s Disease International, London. Anoop A, Singh PK, Jacob RS, Maji SK (2010) CSF Biomarkers for Alzheimer’s Disease Diagnosis. Int J Alzheimers Dis 2010, 1-12. Hampel H, B¨urger K, Teipel SJ, Bokde ALW, Zetterberg H, Blennow K (2008) Core candidate neurochemical and imaging biomarkers of Alzheimer’s disease. Alzheimer Dement 4, 38-48. Blennow K, Hampel H (2003) CSF markers for incipient Alzheimer’s disease. Lancet Neurol 2, 605-613. Lopez OL, Becker JT, Wahed AS, Saxton J, Sweet RA, Wolk DA, Klunk W, Dekosky ST (2009) Long-term effects of the
or P
528
ACKNOWLEDGMENTS
uth
527
[20] [21]
[22]
[23]
[24]
[25] [26]
11
585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649
656 657
[28]
658 659 660 661
[29]
662 663 664 665 666 667 668
[30]
669 670 671 672
[31]
673 674 675 676
[32] [33]
677 678 679
[34]
680 681 682
[35]
683 684 685
[36]
686 687 688 689 690 691 692 693 694 695 696
[37]
[40]
[41]
[42]
roo f
655
or P
654
[39]
Gasper MC, Ott BR, Lapane KL (2005) Is donepezil therapy associated with reduced mortality in nursing home residents with dementia? Am J Geriatr Pharmacother 3, 1-7. Teipel SJ, Ewers M, Reisig V, Schweikert B, Hampel H, Happich M (2007) Long-term cost-effectiveness of donepezil for the treatment of Alzheimer’s disease. Eur Arch Psychiatry Clin Neurosci 257, 330-336. Bond M, Rogers G, Peters J, Anderson R, Hoyle M, Miners A, Moxham T, Davis S, Thokala P, Wailoo A, Jeffreys M, Hyde (2012) The effectiveness and cost-effectiveness of donepezil, galantamine, rivastigmine and memantine for the treatment of Alzheimer’s disease (review of Technology Appraisal No. 111): A systematic review and economic model. Health Technol Assess 16, 1-470. L´opez-Bastida J, Hart W, Garc´ıa-P´erez L, Linertov´a R (2009) Cost-effectiveness of donepezil in the treatment of mild or moderate Alzheimer’s disease. J Alzheimers Dis 16, 399-407. Kasuya M, Meguro K (2010) Health economic effect of donepezil treatment for CDR 0.5 converters to Alzheimer’s disease as shown by the Markov model. Arch Gerontol Geriatr 50, 295-299. Banerjee S, Wittenberg R (2009) Clinical and cost effectiveness of services for early diagnosis and intervention in dementia. Int J Geriatr Psychiatry 24, 748-754. Handels RL, Aalten P, Wolfs CA, OldeRikkert M, Scheltens P, Visser PJ, Joore MA, Severens JL, Verhey FR (2012) Diagnostic and economic evaluation of new biomarkers for Alzheimer’s disease: The research protocol of a prospective cohort study. BMC Neurol 10, 72. Stewart R (2012) Mild cognitive impairment–the continuing challenge of its “real-world” detection and diagnosis. Arch Med Res 43, 609-614. Mueller SG, Weiner MW, Thal LJ, Petersen RC, Jack C, Jagust W, Trojanowski JQ, Toga AW, Beckett L (2005) The Alzheimer’s disease neuroimaging initiative. Neuroimaging Clin N Am 15, 869-877. Visser PJ, Verhey FRJ, Boada M, Bullock R, De Deyn PP, Frisoni GB, Frolich L, Hampel H, Jolles J, Jones R, Minthon L, Nobili F, Olde Rikkert M, Ousset PJ, Rigaud AS, Scheltens P, Soininen H, Spiru L, Touchon J, Tsolaki M, Vellas B, Wahlund LO, Wilcock G, Winblad B (2008) Development of screening guidelines and clinical criteria for predementia Alzheimer’s disease. The DESCRIPA Study. Neuroepidemiology 30, 254-265.
uth
[27]
[38]
[43]
[44]
dA
653
cte
652
concomitant use of memantine with cholinesterase inhibition in Alzheimer disease. J Neurol Neurosurg Psychiatry 80, 600607. L´opez-Bastida J1, Hart W, Garc´ıa-P´erez L, Linertov´a R (2009) Cost-effectiveness of donepezil in the treatment of mild or moderate Alzheimer’s disease. J Alzheimers Dis 16, 399-407. Tappenden P, Chilcott J, Brennan A, Squires H, Stevenson M (2012) Whole disease modeling to inform resource allocation decisions in cancer: A methodological framework. Value Health 15, 1127-1136. Grupo de trabajo de la Gu´ıa de Pr´actica Cl´ınica sobre la atenci´on integral a las personas con enfermedad de Alzheimer y otras demencias (2010) Gu´ıa de Pr´actica Cl´ınica sobre la atenci´on integral a las personas con enfermedad de Alzheimer y otras demencias. Plan de Calidad para el Sistema Nacional de Salud del Ministerio de Sanidad, Pol´ıtica Social e Igualdad, Madrid. Beach TG, Monsell SE, Phillips LE, Kukull W (2012) Accuracy of the clinical diagnosis of Alzheimer disease at National Institute on Aging Alzheimer Disease Centers, 2005–2010. J Neuropathol Exp Neurol 71, 266-273. Fitzpatrick AL, Kuller LH, Lopez OL, Kawas CH, Jagust W (2005) Survival following dementia onset: Alzheimer’s disease and vascular dementia. J Neurol Sci 15, 43-49. Oblikue Consulting. Base de datos de costes sanitarios Esalud. Consejo General de Colegios Oficiales de Farmac´euticos. Base de Datos del Conocimiento Sanitario (Bot PLUS) [online], https://botplusweb.portalfarma.com/ Briggs AH, Claxton K, Sculpher MJ (2006) Decision Modelling for Health Economic Evaluation, Oxford University Press, New York. Fenwick E, Claxton K, Sculpher M (2001) Representing uncertainty: The role of cost-effectiveness acceptability curves. Health Econ 10, 779-787. Calv´o-perxas L, Osuna MT, Gich J, Eligio-Hernandez E, Linares M, Vinas M, Casas I, Turro-Garriga O, LopezPousa S, Garre-Olmo J (2012) Caracter´ısticas cl´ınicas y demogr´aficas de los casos de demencia diagnosticados en la Regi´on Sanitaria de Girona durante el per´ıodo 2007–2010: Datos del Registro de Demencias de Girona (ReDeGi). Rev Neurol 54, 399-406. Bolet´ın Oficial de Canarias n´um. 146. ORDEN de 5 de julio de 2012, por la que se introducen nuevas prestaciones y se modifica la cuant´ıa de los precios p´ublicos de los servicios sanitarios prestados por el Servicio Canario de la Salud, http://www.gobiernodecanarias.org/boc/2012/146/
rre
651
co
650
C.V. Nazco et al. / Cost-Effectiveness of the Use of Biomarkers in Cerebrospinal Fluid
Un
12
[45]
[46]
[47]
697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740
Journal of Alzheimer’s Disease xx (20xx) x–xx DOI 10.3233/JAD-132216 IOS Press
3
roo f
2
Cost-Effectiveness of the Use of Biomarkers in Cerebrospinal Fluid for Alzheimer’s Disease
or P
1
5
Cristina Valc´arcel Nazcoa,b,∗ , Lilisbeth Perestelo P´ereza,c , Jos´e Luis Molinuevod , Javier Marc,e , Iv´an Castillaa,f and Pedro Serrano Aguilara,c
6
a Evaluation
7
b Canary
4
12
Handling Associate Editor: Ramon Luengo-Fernandez
9 10
cte
Accepted 17 April 2014
dA
8
uth
11
Unit of the Canary Islands Health Service (SESCS), Tenerife, Spain Islands Foundation for Health and Research (FUNCIS), Tenerife, Spain c Health Services Research on Chronic Patients Network (REDISSEC), Bizkaia, Spain d Alzheimer’s disease and other cognitive disorders unit. IDIBAPS, Hospital Clinic, Barcelona, Spain e Clinical Management Service, Alto Deba Hospital, Mondragon, Spain f Spanish Consortium for Research on Epidemiology and Public Health (CIBERESP), Madrid, Spain
31
Keywords: Alzheimer’s disease, amyloid-, biomarkers, cost-effectiveness, diagnosis, phosphorylated tau, total tau
16 17 18 19 20 21 22 23 24 25 26 27 28 29
co
15
Un
14
rre
30
Abstract. Background: The use of cerebrospinal fluid (CSF) biomarkers could facilitate early detection of Alzheimer’s disease (AD) in patients with mild cognitive impairment (MCI) and the differential diagnosis between AD and non-AD dementias. The aim of this study is to assess the clinical and economic value of CSF biomarkers to diagnose AD. Objective: To determine the cost-effectiveness of the use of amyloid- peptide (A42 ), total tau, and phosphorylated tau proteins in CSF to diagnose AD in MCI and dementia patients. Methods: An economic evaluation was performed by means of cost-effectiveness analysis comparing two AD diagnostic alternatives: the combined determination of A42 proteins, total tau, and phosphorylated tau in CSF as biomarkers of AD, and the standard clinical diagnosis based on the Related Disorders Association (NINDS-ADRDA) criteria. A decision analytic model was developed to synthesize the identified evidence and to compare the costs and effectiveness associated with each diagnostic strategy. A probabilistic sensitivity analysis using 2nd order Monte Carlo simulations was performed. Subsequently, acceptability curves were calculated and ANCOVA models were applied to the results of the Monte Carlo simulations in order to identify the parameters that led greater variability in the model outcomes. Results: The use of CSF biomarkers as an early diagnostic strategy of AD in MCI patients is a dominant alternative (less costly and more effective strategy than diagnostic criteria). In dementia patients, although there is a higher uncertainty, biomarkers in CSF seem a more cost-effective alternative than diagnostic criteria. Conclusions: Detecting AD in MCI patients by determining A42 , total tau, and phosphorylated tau proteins biomarkers in CSF is a cost-effective diagnostic alternative. No conclusive results were obtained on dementia patients.
13
∗ Correspondence
to: Cristina Valc´arcel Nazco, Servicio de Evaluaci´on del Servicio Canario de la Salud (SESCS), Camino Candelaria, 44, 38109 El Rosario, Santa Cruz de Tenerife, Canary Islands, Spain. Tel.: +34 922 68 40 19; Email: cristina.valcar [email protected].
ISSN 1387-2877/14/$27.50 © 2014 – IOS Press and the authors. All rights reserved
39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82
roo f
38
or P
37
in research; however, they have now fallen behind the unprecedented growth of scientific knowledge. For this reason, in 2007, the International Working Group (IWG) new research criteria [15] introduced a new conceptualization of AD and created a framework to establish a specific diagnosis in the pre-dementia or prodromal stage of the disease [16]. There is increasing consensus to understand AD as a clinical-biological entity, in which biomarkers, especially pathophysiological markers revealing the underlying pathology, represent the biological counterpart of the diagnosis, and specific symptoms, such as episodic memory deficits, account for the clinical one. In 2011, the National Institute of Aging Alzheimer Association (NIA-AA) workgroups also published biomarker supported diagnostic criteria to cover all disease stages [17]. These new criteria, which allow different level of diagnostic certainty based on biomarker results, permit detecting the asymptomatic or preclinical stage [18], the predementia or prodromal stage, termed MCI due to AD [19], and the dementia stage, named dementia due to AD [7]. However, the diagnosis in the predementia stage, applying the IWG criteria [15] or the MCI due to AD criteria [19], are considered to be applicable only in the research setting, since although some research supports the notion that biomarkers may help in predicting conversion to dementia [20, 21], there is yet no clear conviction that the time of AD diagnosis can be reliably advanced [22]. The appropriate selection, detection and visualization of biomarkers can aid in the diagnosis of AD and may prevent a significant number of false positive diagnoses by clinicians using clinical guidelines alone. Although significant advances have been made in the field of neuroimaging, biomarkers based on CSF are at present the most convenient for studying disease progression [11, 23, 24]. Among the better studied biomarkers are the quantification of the amyloid- peptide (A42 ) and tau protein (total tau and phosphorylated tau proteins) in the CSF. Accumulation of amyloid plaques and neurofibrillary tangles probably starts 20 to 30 years before the clinical onset of the disease. Therefore, CSF biomarkers are the most suitable candidates to facilitate AD diagnosis in the very early stages of the disease, long before symptoms onset. Moreover, since it may be optimal to treat the neuropathology as early as possible, biomarkers of preclinical AD are likely to play a pivotal role in the development of the next generation of therapies [24]. The diagnostic sensitivity and specificity are between 80–90% for each biomarker separately [25]. Combining the results of all three biomarkers is now
dA
36
cte
35
rre
34
Alzheimer’s disease (AD) is the leading cause of dementia in the elderly [1] involving up to 70% of dementia cases. The reported prevalence in Europe is 6.2% for people over 60 years old and it is estimated a 40% increase in number of people with AD by 2030 [2]. The health and social relevance of AD and the need of effective treatments have generated an increased interest on the development and evaluation of valid diagnostic procedures for its early detection. To date, AD is diagnosed at the dementia stage of the disease when cognitive impairment interferes with the patient’s functional capacities for daily living [3]. This clinical threshold does vary among individuals, depending on factors such as premorbid cognitive performance, level of intelligence, complexity of everyday activities, or level of perception of the informant. Currently, the standard diagnostic procedure consists of a set of internationally validated clinical criteria. The most frequently used criteria are those defined in the 4th Diagnostic and Statistical Manual of Mental Disorders (DSM-IV-TR) [4, 5], the 10th International Classification of Disease (ICD-10) [6], and in the guidelines of the National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Association (NINDS-ADRDA guidelines) [3, 7]. These criteria are based on the detection of the dementia syndrome and the classic features of AD, being more oriented to the exclusion of other non-degenerative causes of dementia. Although diagnosis based on these criteria has proven to be very sensitive to AD when compared to other dementias, its specificity reaches only 50 to 60% [8]. Hence, using only clinical criteria to diagnose AD may lead to late, and unnecessary and expensive treatment due to this substantial number of false positives. An earlier and more specific diagnosis may be of help to patients and clinicians, regardless that the underlying biological process cannot be stopped [9], by contributing to guiding therapy and properly advising patients and families. These reasons explain the research interest in developing valid neuropsychological, neuroimaging, blood, or cerebrospinal fluid (CSF) markers that could allow both the early detection of AD in patients with mild cognitive impairment (MCI) [10–13] and the differential diagnosis between AD and non-AD dementias [10, 12, 14]. The NINCDS–ADRDA and the DSM-IV-TR criteria for AD are the prevailing diagnostic standards
co
33
INTRODUCTION
Un
32
C.V. Nazco et al. / Cost-Effectiveness of the Use of Biomarkers in Cerebrospinal Fluid
uth
2
83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134
C.V. Nazco et al. / Cost-Effectiveness of the Use of Biomarkers in Cerebrospinal Fluid
140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159
160
161
METHODS Literature search
172
Model overview
165 166 167 168 169 170
173 174 175 176 177 178 179 180 181
co
164
Un
163
rre
171
A literature search was performed, without temporal restrictions, to identify systematic reviews, meta-analysis, and economic evaluations on the diagnostic performance of CSF biomarkers of AD, in the databases MEDLINE and PREMEDLINE, Centre for Reviews and Dissemination and EMBASE. The following search terms were used: (“alzheimer’s disease diagnosis” or “alzheimer’s disease”) and (“abeta-42” or “T-tau” or “P-tau” or “tau” or “phospho-tau” or “phosphorylated tau”).
162
roo f
139
or P
138
Figure 1 shows a schematic outline of the modeling approach. The model assesses costs from the perspective of the Spanish National Health System (NHS) expressed in Euros (D ) at a 2013 base price. The principal effectiveness measure used was “appropriate diagnosis” that would enable correct information to apply symptomatic treatment with donepezil. Several studies have shown that acetylcholinesterase inhibitors like donepezil may reduce costs and even delay institutionalization but do not have any impact on patient survival [26]. Therefore, no QALYs were calculated because donepezil does not modify the natural history of AD. Consequently, the health outcomes of the two alternatives were considered to be the same. It should be noted that, in the diagnostic criteria alternative, many patients with dementias not associated with AD are treated for AD and do not obtain any benefit. Several trials have revealed the benefits of acetylcholinesterase inhibitors, such as donepezil on cognition, activities of daily living, and global functioning. Additionally, the cost-effectiveness of this treatment has been proved in economic studies [27]. Results were obtained separately for two clinical scenarios. The first scenario (Scenario 1) considered a hypothetic cohort of patients aged 60 and older with MCI; the second scenario (Scenario 2) assumed a hypothetic cohort of patients aged 60 and older, with symptoms of dementia. A lifetime horizon was used in the analysis, and costs were discounted at 3%. The model was implemented in TreeAge Pro 2009 Healthcare (TreeAge Software, Inc., USA).
uth
137
considered by some authors as the best biomarker of AD [24]. Though the diagnostic sensitivity of its combined determination is very high, its specificity is limited, as they might be also altered in other neurodegenerative diseases [14]. There are several misunderstandings in how biomarkers should be interpreted from a clinical perspective, in how the new research criteria could be applied in clinical practice, and in the ethical background supporting early diagnosis. This existing uncertainty demands an urgent need for health technology assessment to address the evaluation of the diagnostic strategies including biomarkers for AD compared to current clinical practice. Therefore, the aim of this study is to assess the clinical and economic value of CSF biomarkers for the early diagnosis and detection of AD; being the main objective to determine the cost-effectiveness of the use of A42 , total tau, and phosphorylated tau proteins in CSF for early detection of AD and confirmation of the diagnosis. The objective of our study was to determine the cost-effectiveness of the use of A42 , total tau, and phosphorylated tau proteins in CSF to diagnose AD in MCI and dementia patients.
dA
136
cte
135
3
An economic evaluation was performed by means of cost-effectiveness analysis comparing two diagnostic alternatives for AD: the combined determination of A42 , total tau, and phosphorylated tau proteins in CSF as biomarkers of AD and the standard clinical diagnosis based on the NINDS-ADRDA criteria [3]. A decision analytic model was developed to synthesize the evidence identified and to compare the costs and effectiveness associated with each technology.
Model inputs Parameters Model inputs were estimated from the literature [11, 12, 29, 30]. Table 1 shows the values and information sources of the most important parameters used in the base case for each scenario. In the first scenario, the cost-effectiveness analysis comprehends that, by definition, MCI patients cannot be diagnosed of AD at the time of the first evaluation and analysis; therefore the probability of detecting a suspected case (see below) in this scenario implicitly implies the probability of a given patient to develop AD in the future. To estimate the probability of detecting a suspected case of AD in each arm of the model, we proceeded as follows: being P AD the “Probability of developing AD”, P suspected AD the “Probability of detecting
182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215
216
217 218 219 220 221 222 223 224 225 226 227 228 229 230 231
C.V. Nazco et al. / Cost-Effectiveness of the Use of Biomarkers in Cerebrospinal Fluid
or P
roo f
4
uth
Fig. 1. Decision tree model structure. P suspected AD biomarkers = Probability of suspected case of AD (biomarkers). P suspected AD diagnostic criteria = Probability of suspected case of AD (diagnostic criteria). PPV biomarkers = Positive predictive value of biomarker strategy. NPV biomarkers = Negative predictive value of biomarker strategy. PPV diagnostic criteria = Positive predictive value of diagnostic criteria. NPV diagnostic criteria = Negative predictive value of diagnostic criteria. #=complementary probability.
Table 1 Base case estimates
Item
Un
co
rre
cte
SCENARIO 1 (MCI patients) AD incidence AD prevalence Sensitivity of biomarkers Specificity of biomarkers Sensitivity of diagnostic criteria Specificity of diagnostic criteria Probability of suspected case of AD (biomarkers) Probability of suspected case of AD (diagnostic criteria) SCENARIO 2 (Dementia patients) AD incidence AD prevalence Sensitivity of biomarkers Specificity of biomarkers Sensitivity of diagnostic criteria Specificity of diagnostic criteria Probability of suspected case of AD (biomarkers) Probability of suspected case of AD (diagnostic criteria)
dA
Item
BIOMARKERS First neurology outpatient appointment Lumbar puncture Reactive per sample (ELISA kits) Hour worked of laboratory technician Hour worked of physician Subsequent neurology outpatient appointment Treatment with donepezil during lifetime DIAGNOSTIC CRITERIA First neurology outpatient appointment Subsequent neurology outpatient appointment Treatment with donepezil during lifetime
Estimate
Source
0.0319 0.086 0.81 0.87 0.87 0.58 0.1517 0.4344
[29] [29] [11] [11] [30] [30] [11, 29] [29, 30]
0.0408 0.1226 0.84 0.71 0.873 0.443 0.3124 0.5699
[30] [30] [12] [12] [30] [30] [12, 29] [29, 30]
Table 2 Resource use and unit costs (2013 D ) Resource use
Unit cost
Source
1 1 3 1 1 1 10 mg daily
D 170.82 D 158.28 D 22.08 D 7.99 D 16.55 D 102.49 D 2.33 (annual cost of D 850.06)
[32] [32] ELISA kits distributor CHS CHS [32] [33]
1 1 10 mg daily
D 170.82 D 102.49 D 2.33 (annual cost of D 850.06)
[32] [32] [33]
ELISA, Enzyme-linked immunosorbent assay; CHS, Canarian Health Service.
C.V. Nazco et al. / Cost-Effectiveness of the Use of Biomarkers in Cerebrospinal Fluid
236
237
238
P AD = P suspected AD∗ P TP ∗
+(1 − P suspected AD) (1 − P TN)
239
244 245 246
247
248
249
250
251 252 253
254 255 256 257 258 259 260
261 262 263 264 265 266 267 268 269 270 271 272 273 274 275
PPV = Sens ∗ Prev/Sens ∗ Prev + (1 − Esp) ∗ (1 − Prev) NPV = Esp ∗ (1 − Prev)/Esp ∗ (1 − Prev) + (1 − Sens) ∗ Prev where Sens = Sensibility of the diagnostic strategy, Esp = Specificity of the diagnostic strategy, and Prev = Prevalence of AD.
A sensitivity analysis was conducted in order to determine whether the results of the study changed when the values of certain variables were altered. The difference in cost between strategies was divided by the difference in effects to measure the cost per case detected, expressed as the incremental costeffectiveness ratio (ICER). In order to account for the uncertainty in key model parameters, deterministic and probabilistic sensitivity analysis using 10,000 Monte Carlo simulations [34] were performed. One suggested method to represent this uncertainty is costeffectiveness acceptability curves [35]. Acceptability curves represent the probability that a given intervention is cost-effective at different values of willingness to pay (WTP) per effectiveness unit. Table 3 shows the parameters used in the sensitivity analysis [11, 12, 24, 29, 30, 36, 37]. The cost-effectiveness of an alternative depends on the WTP of the payer for each additional effectiveness unit. Since no references have been found for the Spanish NHS WTP, results have been presented for WTP of 10,000 and 1,000 D for correctly diagnosed case. Finally, the percentage variability of the costs and effects explained by each parameter of the model was calculated using ANCOVA models. The advantage of this method compared to a non-probabilistic sensitivity analysis is that, for each parameter set, variations in the other model parameters are taken into account.
cte
243
Clinical effectiveness Effectiveness was measured in natural clinical units in terms of correctly diagnosed cases of AD. Therefore, the measure of effectiveness in each arm of the model is directly related to the sensitivity and specificity of each diagnostic strategy, that is, true positives and true negatives of each alternative compared.
rre
242
co
241
Thus, eliminating the unknown P suspected AD, we obtained the probability of detecting a suspected case in each diagnostic strategy based on sensitivity and specificity. The positive and negative predictive values (PPV and NPV) of each diagnostic strategy were estimated, respectively, by the formulae:
Resource use and unit costs According to the perspective adopted for the analysis, only direct health care costs such as drugs (donepezil), medical visits and diagnostic procedures, were considered. As mentioned before, treatment with donepezil was assumed to have no influence on patient survival, so estimates of time to death in the model were assumed to be identical for treated and untreated patients. Median survival with AD was estimated at 7.1 years (95% CI 6.7–7.5 years) [31] according to the literature. Treatment with donepezil continues right up until death. Table 2 shows resource use per patient and unit costs used in the analysis [32, 33].
Un
240
276
roo f
235
Sensitivity analysis
or P
234
a suspected case of AD”, P TP the “Probability of obtaining a True Positive (sensitivity of diagnostic strategy)”, and P TN the “Probability of obtaining a True Negative (specificity of diagnostic strategy)”, then
uth
233
dA
232
5
277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305
RESULTS
306
Base case results
307
Scenario 1 (patients with MCI) shows a lower average cost per patient for biomarkers use than for conventional diagnostic clinical criteria (1,336.06 D versus 3,167.39 D , respectively). This is due to the fact that conventional clinical criteria misdiagnosed more cases of AD treated with donepezil. In addition, biomarkers in CSF have a likelihood of accurate detection of 88,73%, which means that most MCI patients with a positive biomarker profile will develop AD in the future, whereas this probability falls below 52% for the alternative strategy. Base case results for MCI patients reveal that CSF biomarkers for early AD detection are a dominant diagnostic strategy. The average cost of a dementia patient (Scenario 2) diagnosed by CSF biomarkers is higher than the average cost of a dementia patient diagnosed
308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323
6
C.V. Nazco et al. / Cost-Effectiveness of the Use of Biomarkers in Cerebrospinal Fluid Table 3 Parameters used in sensitivity analysis Distribution (Mean ± Error)
SCENARIO 1 (MCI patients) AD incidence AD prevalence Sensitivity of biomarkers Specificity of biomarkers Sensitivity of diagnostic criteria Specificity of diagnostic criteria SCENARIO 2 (Dementia patients) AD incidence AD prevalence Sensitivity of biomarkers Specificity of biomarkers Sensitivity of diagnostic criteria Specificity of diagnostic criteria BOTH SCENARIOS Cost of first neurology outpatient appointment Cost of subsequent neurology outpatient appointment Cost, per hour, of a physician Cost, per hour, of a laboratory technician Cost of lumbar puncture Discount Annual cost of donepezil
BETA (0.0319 ± 0.02) BETA (0.086 ± 0.02) BETA (0.81 ± 0.081) BETA (0.87 ± 0.043) BETA (0.859 ± 0.085) BETA (0.471 ± 0.047)
or P
BETA (0.0408 ± 0.028) BETA (0.1226 ± 0.03) BETA (0.84 ± 0.071) BETA (0.71 ± 0.106) BETA (0.873 ± 0.087) BETA (0.443 ± 0.04)
329
cte
328
rre
327
[29, 30] [29] [12] [12] [30] [30] [32, 37] [32, 37] CHS CHS [32] 0%–3% D 680.048–D 850.06
Results of sensitivity analysis
330
The results of the deterministic sensitivity analysis are shown in Table 4. In all cases, the use of CSF biomarkers resulted a dominant strategy, i.e., less costly and more effective than the alternative strategy.
Table 4 Results of deterministic sensitivity analysis
Parameter SCENARIO 1 (MCI patients) Discount Annual cost of donepezil SCENARIO 2 (Dementia patients) Discount Annual cost of donepezil
co
326
by conventional clinical criteria (2,198.33 D versus 3,330.28 D , respectively). In this scenario, biomarkers in CSF have a likelihood of accurate detection of 75.66%; whereas the probability of detecting a suspected case of AD in the alternative strategy is 52%.
Parameter SCENARIO 1 (MCI patients) Cost Accurate diagnosis∗ SCENARIO 2 (Dementia patients) Cost Accurate diagnosis∗ ∗ Probability
Un
325
[29] [29] [11] [11] [30] [30]
dA
uth
GAMMA (170.82 ± 42.705) GAMMA (102.49 ± 25.62) GAMMA (16.55 ± 4.965) GAMMA (7.99 ± 2.397) GAMMA (158.28 ± 47.48)
CHS, Canarian Health Service.
324
Source
roo f
Item
Base case
Value
ICER
3% D 850.06
0% D 680.048
Dominant Dominant
3% D 850.06
0% D 680.048
Dominant Dominant
Table 5 Results of probabilistic sensitivity analysis
Biomarkers
Diagnostic criteria
Incremental
ICER
1,334.30 D [906.79; 1,871.62] 0.8857 [0.7965; 0.9506]
3,166.96 D [2,672.22; 3,649.06] 0.5184 [0.4234; 0.6144]
D −1,832.65 [−2,473.48; −1,142.65] 0.3672 [0.2418; 0.4833]
Dominant
2,198.05 D [1,256.53; 3,346.87] 0.7541 [0.5487; 0.9119]
3,331.87 D [2,854.64; 3,790.06] 0.5149 [0.4169; 0.6170]
−1,133.82 D [−2,176.82; 77,95] 0.2392 [0.0216; 0.4215]
Dominant [Dominant; 3,607.54]
of achieving the right diagnosis.
331 332 333 334
7
Un
co
rre
cte
dA
uth
or P
roo f
C.V. Nazco et al. / Cost-Effectiveness of the Use of Biomarkers in Cerebrospinal Fluid
Fig. 2. Incremental cost-effectiveness plane.
335 336 337 338 339 340
Table 5 presents the results of the probabilistic sensitivity analysis for the two alternative scenarios. The results from scenario 1 suggest similar conclusions to the base case analysis. The use of CSF biomarkers as an early diagnostic strategy of AD in patients with MCI is a dominant alternative. In scenario 2, the average ICER
value highlights the uncertainty over the decision to take. Figure 2 shows the incremental cost-effectiveness plane for each considered scenario. These graphs represent the cost and effectiveness pairs obtained from each simulation model, and the average ICER. As seen
341 342 343 344 345 346
C.V. Nazco et al. / Cost-Effectiveness of the Use of Biomarkers in Cerebrospinal Fluid
uth
or P
roo f
8
Fig. 3. Acceptability curve. scenario 2: Dementia patients.
370
DISCUSSION
352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368
371 372 373 374 375 376
cte
351
rre
350
co
349
Un
348
cial interventions [43]. However, we have not been able to identify studies on the cost-effectiveness of diagnostic procedures focused on early detection of AD when only signs of MCI are visible. This study has developed an economic evaluation that compares, in terms of diagnostic efficiency, two alternatives for the detection of AD. We have considered two scenarios with different patients: patients with MCI who have no symptoms of dementia, and patients with those symptoms. Although there are not available effective drugs to change AD natural history, symptomatic drug treatment is widely used according to clinical criteria. The use of biomarkers in CSF for early and more valid diagnosis of AD plays a leading role as it contributes to an accurate and more efficient diagnosis by means of objective and reliable procedures. These findings could be of interest for clinical research, to inform patients and families, and to guide earlier treatment decisions when more effective alternatives exist. This economic evaluation does not analyze the efficiency of diagnosis of MCI leading to AD or the early diagnosis of AD in terms of health-related quality of life (HRQOL). Instead, this economic evaluation is interested in the diagnostic efficiency of biomarkers for the detection of AD, using “cases detected and treated correctly” as the measure of effectiveness. Although using “cases detected and treated correctly” has the limitation of being an intermediate outcome, it was selected because the objective of this report was to assess the diagnostic efficiency of protein biomarkers in CSF.
dA
369
in Fig. 2, the use of biomarkers in CSF is less costly and more effective than diagnostic criteria in MCI patients. For dementia patients, a confidence interval for ICER of [Dominant; 3,607.54] was obtained. The acceptability curve of Scenario 2 for different WTP values is presented in Fig. 3. If we draw a vertical line on the graph at a given WTP value, the curve represents the probability that the use of biomarkers in CSF is a cost-effective alternative. Therefore, in this scenario (dementia patients), the probability of accepting biomarkers as a cost-effective alternative is approximately 80% for 1,000 D WTP. The acceptability curve for scenario 1 is not presented because the biomarker strategy is a dominant alternative. The ANCOVA analysis results are shown in Fig. 4. In scenario 1, the parameter that has the greatest impact on both costs and effectiveness is the specificity of the conventional clinical diagnostic criteria. Specificity of biomarker strategy also has a remarkable impact on both the cost and effectiveness variability. When focusing on scenario 2, the parameter that has a greater impact on both costs and effects is the specificity of biomarkers in CSF.
347
According to our results, the use of A42 , total tau, and phosphorylated tau proteins tests in CSF is clearly cost-effective in MCI patients but is not conclusive in dementia patients. Previous economics evaluations reported on the cost-effectiveness of AD treatments, including both pharmacologic [38–42] and psychoso-
377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408
9
Un
co
rre
cte
dA
uth
or P
roo f
C.V. Nazco et al. / Cost-Effectiveness of the Use of Biomarkers in Cerebrospinal Fluid
409 410 411 412 413 414 415 416 417
Fig. 4. Results of ANCOVA analysis.
To the best of our knowledge, this is the first study to analyze the diagnostic efficiency of protein biomarkers in CSF as compared to routine clinical diagnosis of AD. Previously, a study published a protocol designed to conduct an economic evaluation of these biomarkers in CSF to improve HRQOL in the hypothetic context of effective drugs being developed [44]. The results of this study will be published in two years and they will be theoretic.
The results of this economic evaluation prove the potential cost-effectiveness of the combined use of AD CSF biomarkers against the application of standard clinical criteria in patients with MCI who will develop AD later. For this group of patients, a dominant ICER was obtained. According to these results, the use of biomarkers in CSF is a cost-effective diagnostic alternative. The very low error probabilities given by the acceptability curves in patients with MCI, makes
418 419 420 421 422 423 424 425 426
C.V. Nazco et al. / Cost-Effectiveness of the Use of Biomarkers in Cerebrospinal Fluid
434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473
476 477
cte
433
rre
432
co
431
Un
430
roo f
Limitations
429
and sensitivity of the standard clinical diagnostic criteria for accurate diagnosis of AD. The uncertainty surrounding these parameter estimates influences the inability to determine which diagnostic strategy is more efficient for dementia patients. Another limitation to consider relates to the measure of effectiveness selected. In general, the outcome “cases detected and treated correctly” is considered an intermediate step in economic analysis. As mentioned above, no QALYs were calculated because donepezil does not modify the natural history of AD given that several studies have shown that acetylcholinesterase inhibitors do not have any impact on patient survival [26]. Moreover, this measure responds to the primary purpose of this economic evaluation of studying the diagnostic efficiency of protein biomarkers in CSF. Therefore, the model do not take into account variations in disease progression, as this would be the same in both strategies, but it does consider the cost savings related with avoided unnecessary drug treatment to patients wrongly diagnosed. The exclusion of other costs related to institutionalization delay could be considered a limitation. However, as biomarkers diagnostic strategy is dominant in MCI patients, the addition of these costs would reaffirm the conclusions of our study. Last but not least, the absence of a threshold value for the WTP in the Spanish context adds extra uncertainty to the conclusions of this study. This limitation is overcome in this study by presenting the results for different values of hypothetic WTP thresholds. According to the WTP taken as a reference, the results may favor or go against the efficiency of protein biomarkers in CSF as a diagnostic strategy for patients with symptoms of dementia.
or P
475
428
dA
474
CSF biomarkers a cost-effective procedure compared to conventional clinical diagnostic criteria. Variations in the specificity of clinical diagnostic criteria and in the specificity of biomarkers may, however, alter these results because these parameters have a large impact on model results. However, neither the prevalence of AD (nor incidence) nor biomarkers sensitivity affect the results of this model. The results obtained in MCI patients entail significant challenges for research on the diagnostic and therapeutic process of the early stages of AD. Although exact figures on the proportion of people with MCI who may progress to dementia are not known, there is consensus on the fact that MCI patients have a higher incidence of dementia than the general population [19, 45]. For patients with symptoms of dementia, the uncertainty around the parameters of the economic model developed is high. For this group of patients with suspected AD, the results obtained do not enable us to assure that biomarkers in CSF are more cost-effective than standard diagnostic criteria in terms of diagnostic efficiency. The parameters with more influence on the economic evaluation for dementia patients are biomarkers specificity and the cost of lumbar puncture. Precisely, the cost of lumbar puncture is highly variable among Spanish regions. This is an important element to be taken into account should standardization of this diagnostic technique on a national scale become an objective. The results of this economic evaluation may help initiatives such as The Alzheimer’s Disease Neuroimaging Initiative (ADNI), which attempts to identify and integrate information on diagnostic procedures based on neuroimaging and biomarkers [46]. Our results may also be useful for studies such as the Development of Screening Guidelines and Clinical Criteria for Predementia AD (DESCRIPA), which aims to develop guidelines for screening AD pre-dementia states in the general population [47]. Therefore, this new technology could be cost-effective in MCI patients from the perspective of the validity and efficiency of new diagnostic procedures to facilitate clinical and evaluative research on the effectiveness of new therapeutic interventions, in addition to informing patients and families and thereby assisting their earlier overall arrangements and, hopefully, a shared decision making process.
427
uth
10
The main limitation of this economic evaluation was the difficulty to find quality estimates of the specificity
CONCLUSIONS According to the results of this economic evaluation, detecting AD in patients with MCI by the determination of A42 proteins, total tau, and phosphorylated tau in CSF as biomarkers is a cost-effective alternative from the perspective of the Spanish NHS. However, since there is no defined WTP in Spain for this unit of effectiveness, these results should be treated with caution. When applying the same technique to patients with symptoms of dementia, there is a higher uncertainty in the results. Nevertheless, biomarkers in CSF still seem a more cost-effective alternative than diagnostic criteria.
478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512
513
514 515 516 517 518 519 520 521 522 523 524 525 526
C.V. Nazco et al. / Cost-Effectiveness of the Use of Biomarkers in Cerebrospinal Fluid
530 531 532 533 534 535 536
[14]
[15]
[16]
540
[2]
541 542 543
[3]
544 545 546 547 548
[4]
549 550 551
[5]
552 553
[6]
554 555 556
[7]
558 559 560 561 562 563 564
[8]
565 566 567 568 569
[9]
570 571 572 573
[10]
574 575 576 577
[11]
578 579 580 581 582 583 584
[18]
[19]
rre
557
[17]
dA
539
Khachaturian ZS (1985) Diagnosis of Alzheimer’s disease. Arch Neurol 42, 1097-1105. Alzheimer Disease, International (2010) World Alzheimer Report 2010: The Global Economic Impact of Dementia, Alzheimer Disease International, London. McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM (1984) Clinical diagnosis of Alzheimer’s disease: Report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology 34, 939-944. American Psychiatric Association (2000) Diagnostic and Statistical Manual of Mental Disorders, 4th edition. American Psychiatric Association, Washington DC. Lopez Ibor J (2005) DSM-IV-TR. Manual Diagnostico y Estadistico de los Trastornos Mentales, Masson, Barcelona. World Health, Organization (1992) The ICD-10 Classification of Mental and Behavioural Disorders, World Health Organization, Geneva. McKhann GM, Knopman DS, Chertkow H, Hyman BT, Jack CR, Kawas CH, Klunk WE, Koroshetz WJ, Manly JJ, Mayeux R, Mohs RC, Morris JC, Rossor MN, Scheltens P, Carrillo MC, Thies B, Weintraub S, Phelps CH (2011) The diagnosis of dementia due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7, 263-269. Lim A, Tsuang D, Kukull W, Nochlin D, Leverenz J, McCormick W, Bowen J, Teri L, Thompson J, Peskind ER, Raskind M, Larson EB (1999) Clinico-neuropathological correlation of Alzheimer’s disease in a community-based case series. J Am Geriatr Soc 47, 564-569. Molinuevo JL, Berthier ML, Rami L (2011) Donepezil provides greater benefits in mild compared to moderate Alzheimer’s disease: Implications for early diagnosis and treatment. Arch Gerontol Geriatr 52, 18-22. Mitchell AJ (2009) CSF phosphorylated tau in the diagnosis and prognosis of mild cognitive impairment and Alzheimer’s disease: A meta-analysis of 51 studies. J Neurol Neurosurg Psychiatry 80, 966-975. Monge-Argil´es JA, S´anchez-Pay´a J, Mu˜noz-Ruiz C, Pampliega-P´erez A, Montoya-Guti´errez J, Leiva-Santana C (2010) Biomarcadores en el l´ıquido cefalorraqu´ıdeo de pacientes con deterioro cognitivo leve: Metaan´alisis de su capacidad predictiva para el diagn´ostico de la enfermedad de Alzheimer. Rev Neurol 50, 193-200. Bloudek LM, Spackman DE, Blankenburg M, Sullivan SD (2011) Review and meta-analysis of biomarkers and diag-
cte
[1]
[12]
co
538
REFERENCES
Un
537
roo f
529
The authors gratefully acknowledge the contribution of Cristobal Carnero and Norberto Rodr´ıguez in the design of the economic evaluation model. The authors also thank Laura Vallejo Torres for the review of the manuscript. This study was funded by Spanish Health Ministry (Ministerio de Sanidad, Servicios Sociales e Igualdad). Authors’ disclosures available online (http://www.jalz.com/disclosures/view.php?id=2284).
[13]
nostic imaging in Alzheimer’s disease. J Alzheimers Dis 26, 627-645. Diniz BS, Pinto J´unior JA, Forlenza OV (2008) Do CSF total tau, phosphorylated tau, and beta-amyloid 42 help to predict progression of mild cognitive impairment to Alzheimer’s disease? A systematic review and meta-analysis of the literature. World J Biol Psychiatry 9, 172-182. Van Harten AC, Kester MI, Visser PJ, Blankenstein MA, Pijnenburg YA, van der Flier WM, Scheltens P (2011) Tau and p-tau as CSF biomarkers in dementia: A meta-analysis. Clin Chem Lab Med 49, 353-366. Cummings JL, Dubois B, Molinuevo JL, Scheltens P (2013) International work group criteria for the diagnosis of Alzheimer disease. Med Clin North Am 97, 363-368. Dubois B, Feldman HH, Jacova C, Dekosky ST, BarbergerGateau P, Cummings J, Delacourte A, Galasko D, Gauthier S, Jicha G, Meguro K, O’brien J, Pasquier F, Robert P, Rossor M, Salloway S, Stern Y, Visser PJ, Scheltens P (2007) Research criteria for the diagnosis of Alzheimer’s disease: Revising the NINCDS-ADRDA criteria. Lancet Neurol 6, 734-746. Jack CR, Albert MS, Knopman DS, McKhann GM, Sperling RA, Carrillo MC, Thies B, Phelps CH (2011) Introduction to the recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7, 257-262. Sperling RA, Aisen PS, Beckett LA, Bennett DA, Craft S, Fagan AM, Iwatsubo T, Jack CR Jr, Kaye J, Montine TJ, Park DC, Reiman EM, Rowe CC, Siemers E, Stern Y, Yaffe K, Carrillo MC, Thies B, Morrison-Bogorad M, Wagster MV, Phelps CH (2011) Toward defining the preclinical stages of Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7, 280-292. Albert MS, DeKosky ST, Dickson D, Dubois B, Feldman HH, Fox NC, Gamst A, Holtzman DM, Jagust WJ, Petersen RC, Snyder PJ, Carrillo MC, Thies B, Phelps CH (2011) The diagnosis of mild cognitive impairment due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7, 270-279. Ballard C, Gauthier S, Corbett A, Brayne C, Aarsland D, Jones E (2011) Alzheimer’s disease. Lancet 377, 1019-1031. Buchhave P, Minthon L, Zetterberg H, Wallin AK, Blennow K, Hansson O (2012) Cerebrospinal fluid levels of -amyloid 1-42, but not of tau, are fully changed already 5 to 10 years before the onset of Alzheimer dementia. Arch Gen Psychiatry 69, 98-106. Prince M, Bryce R, Ferri C (2011) World Alzheimer Report 2011: The Benefits of Early Diagnosis and Intervention. Alzheimer’s Disease International, London. Anoop A, Singh PK, Jacob RS, Maji SK (2010) CSF Biomarkers for Alzheimer’s Disease Diagnosis. Int J Alzheimers Dis 2010, 1-12. Hampel H, B¨urger K, Teipel SJ, Bokde ALW, Zetterberg H, Blennow K (2008) Core candidate neurochemical and imaging biomarkers of Alzheimer’s disease. Alzheimer Dement 4, 38-48. Blennow K, Hampel H (2003) CSF markers for incipient Alzheimer’s disease. Lancet Neurol 2, 605-613. Lopez OL, Becker JT, Wahed AS, Saxton J, Sweet RA, Wolk DA, Klunk W, Dekosky ST (2009) Long-term effects of the
or P
528
ACKNOWLEDGMENTS
uth
527
[20] [21]
[22]
[23]
[24]
[25] [26]
11
585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649
656 657
[28]
658 659 660 661
[29]
662 663 664 665 666 667 668
[30]
669 670 671 672
[31]
673 674 675 676
[32] [33]
677 678 679
[34]
680 681 682
[35]
683 684 685
[36]
686 687 688 689 690 691 692 693 694 695 696
[37]
[40]
[41]
[42]
roo f
655
or P
654
[39]
Gasper MC, Ott BR, Lapane KL (2005) Is donepezil therapy associated with reduced mortality in nursing home residents with dementia? Am J Geriatr Pharmacother 3, 1-7. Teipel SJ, Ewers M, Reisig V, Schweikert B, Hampel H, Happich M (2007) Long-term cost-effectiveness of donepezil for the treatment of Alzheimer’s disease. Eur Arch Psychiatry Clin Neurosci 257, 330-336. Bond M, Rogers G, Peters J, Anderson R, Hoyle M, Miners A, Moxham T, Davis S, Thokala P, Wailoo A, Jeffreys M, Hyde (2012) The effectiveness and cost-effectiveness of donepezil, galantamine, rivastigmine and memantine for the treatment of Alzheimer’s disease (review of Technology Appraisal No. 111): A systematic review and economic model. Health Technol Assess 16, 1-470. L´opez-Bastida J, Hart W, Garc´ıa-P´erez L, Linertov´a R (2009) Cost-effectiveness of donepezil in the treatment of mild or moderate Alzheimer’s disease. J Alzheimers Dis 16, 399-407. Kasuya M, Meguro K (2010) Health economic effect of donepezil treatment for CDR 0.5 converters to Alzheimer’s disease as shown by the Markov model. Arch Gerontol Geriatr 50, 295-299. Banerjee S, Wittenberg R (2009) Clinical and cost effectiveness of services for early diagnosis and intervention in dementia. Int J Geriatr Psychiatry 24, 748-754. Handels RL, Aalten P, Wolfs CA, OldeRikkert M, Scheltens P, Visser PJ, Joore MA, Severens JL, Verhey FR (2012) Diagnostic and economic evaluation of new biomarkers for Alzheimer’s disease: The research protocol of a prospective cohort study. BMC Neurol 10, 72. Stewart R (2012) Mild cognitive impairment–the continuing challenge of its “real-world” detection and diagnosis. Arch Med Res 43, 609-614. Mueller SG, Weiner MW, Thal LJ, Petersen RC, Jack C, Jagust W, Trojanowski JQ, Toga AW, Beckett L (2005) The Alzheimer’s disease neuroimaging initiative. Neuroimaging Clin N Am 15, 869-877. Visser PJ, Verhey FRJ, Boada M, Bullock R, De Deyn PP, Frisoni GB, Frolich L, Hampel H, Jolles J, Jones R, Minthon L, Nobili F, Olde Rikkert M, Ousset PJ, Rigaud AS, Scheltens P, Soininen H, Spiru L, Touchon J, Tsolaki M, Vellas B, Wahlund LO, Wilcock G, Winblad B (2008) Development of screening guidelines and clinical criteria for predementia Alzheimer’s disease. The DESCRIPA Study. Neuroepidemiology 30, 254-265.
uth
[27]
[38]
[43]
[44]
dA
653
cte
652
concomitant use of memantine with cholinesterase inhibition in Alzheimer disease. J Neurol Neurosurg Psychiatry 80, 600607. L´opez-Bastida J1, Hart W, Garc´ıa-P´erez L, Linertov´a R (2009) Cost-effectiveness of donepezil in the treatment of mild or moderate Alzheimer’s disease. J Alzheimers Dis 16, 399-407. Tappenden P, Chilcott J, Brennan A, Squires H, Stevenson M (2012) Whole disease modeling to inform resource allocation decisions in cancer: A methodological framework. Value Health 15, 1127-1136. Grupo de trabajo de la Gu´ıa de Pr´actica Cl´ınica sobre la atenci´on integral a las personas con enfermedad de Alzheimer y otras demencias (2010) Gu´ıa de Pr´actica Cl´ınica sobre la atenci´on integral a las personas con enfermedad de Alzheimer y otras demencias. Plan de Calidad para el Sistema Nacional de Salud del Ministerio de Sanidad, Pol´ıtica Social e Igualdad, Madrid. Beach TG, Monsell SE, Phillips LE, Kukull W (2012) Accuracy of the clinical diagnosis of Alzheimer disease at National Institute on Aging Alzheimer Disease Centers, 2005–2010. J Neuropathol Exp Neurol 71, 266-273. Fitzpatrick AL, Kuller LH, Lopez OL, Kawas CH, Jagust W (2005) Survival following dementia onset: Alzheimer’s disease and vascular dementia. J Neurol Sci 15, 43-49. Oblikue Consulting. Base de datos de costes sanitarios Esalud. Consejo General de Colegios Oficiales de Farmac´euticos. Base de Datos del Conocimiento Sanitario (Bot PLUS) [online], https://botplusweb.portalfarma.com/ Briggs AH, Claxton K, Sculpher MJ (2006) Decision Modelling for Health Economic Evaluation, Oxford University Press, New York. Fenwick E, Claxton K, Sculpher M (2001) Representing uncertainty: The role of cost-effectiveness acceptability curves. Health Econ 10, 779-787. Calv´o-perxas L, Osuna MT, Gich J, Eligio-Hernandez E, Linares M, Vinas M, Casas I, Turro-Garriga O, LopezPousa S, Garre-Olmo J (2012) Caracter´ısticas cl´ınicas y demogr´aficas de los casos de demencia diagnosticados en la Regi´on Sanitaria de Girona durante el per´ıodo 2007–2010: Datos del Registro de Demencias de Girona (ReDeGi). Rev Neurol 54, 399-406. Bolet´ın Oficial de Canarias n´um. 146. ORDEN de 5 de julio de 2012, por la que se introducen nuevas prestaciones y se modifica la cuant´ıa de los precios p´ublicos de los servicios sanitarios prestados por el Servicio Canario de la Salud, http://www.gobiernodecanarias.org/boc/2012/146/
rre
651
co
650
C.V. Nazco et al. / Cost-Effectiveness of the Use of Biomarkers in Cerebrospinal Fluid
Un
12
[45]
[46]
[47]
697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740
