- Email: [email protected]
The pitfalls of biomarker-based classification schemes
Alzheimer’s & Dementia - (2017) 1-3
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54
Letter The pitfalls of biomarker-based classification schemes We read with great interest the elegant work by Pascoal et al. [1] describing a synergistic effect between florbetapir uptake values and cerebrospinal fluid (CSF) p-tau levels on the 2-year risk of progression to Alzheimer’s disease (AD) dementia in patients with amnestic mild cognitive impairment (aMCI) from the Alzheimer Disease Neuroimaging Initiative (ADNI). This synergistic toxic effect between amyloid (A) and tau (T) has strong biological evidence and has been shown to accelerate cognitive decline and brain atrophy in healthy elderly [2,3]. The authors used two different biomarker modalities for subject classification. The use of CSF biomarker levels alone would have obvious practical advantages, but might influence the results. Classification schemes using AD biomarkers are highly dependent on the standardization and reproducibility of biomarker measures [4]. A and T can be measured through different biomarkers with similar accuracy [5]. These different biomarkers often correlate with each other, but when classifying a subject into normal or abnormal for a given modality, they sometimes have poor agreement [6]. This has been described as a limitation of classification schemes [4]. To address the effect of the classification method on clinical outcomes in the study of Pascoal et al., we compared the use of CSF Ab1–42 levels instead of florbetapir uptake values and the use of CSF t-tau instead of CSF p-tau values in the same group of aMCI subjects from ADNI. We first assessed the correlation between those biomarkers and their agreement for A and T classification. We then compared the prevalence for each category (A2T2, A1T2, A2T1, and A1T1) and their respective risk of progression to AD dementia in the same group of aMCI. We used the florbetapir uptake values and CSF biomarker levels provided by the
1 Data used in the preparation of this article were obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database (adni.loni. usc.edu). As such, the investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report. A complete listing of ADNI investigators can be found at http://adni.loni.usc.edu/wp-content/ uploads/how_to_apply/ADNI_Acknowledgement_List.pdf. *Corresponding author. Tel.: 134-935565986; Fax:134-935565602. E-mail address: [email protected]
ADNI database and we applied the validated thresholds for each biomarker [7]. Both CSF Ab1–42 levels and the florbetapir uptake values and CSF p-tau and t-tau levels were similarly correlated (r 5 20.695 and 0.636, for A and T, respectively). However, the concordance for subject classification was much lower for T (k 5 0.713 and 0.228 for A and T, respectively). Fig. 1 summarizes the risk of progression to AD dementia and the prevalence of each category when using (1) [18F]florbetapir and p-ptau, (2) CSF Ab1–42 and p-tau, and (3) CSF Ab1–42 and t-tau. In all cases, we replicated the synergistic effects between A and T. However, the group of isolated amyloid pathology (A1T2) emerged as an intermediate risk group (progression rates of 5%–7.1% for the A1T2 group vs. 0% to 1.6% for the A2T2 group at 2 years). These results are in agreement with the study from Vos et al. [8] that described an intermediate progression risk for the isolated amyloid pathology group in a large multicentric MCI cohort after a follow up of 2.7 years. The most important difference was, nonetheless, the dramatic change in the prevalence of the different categories. These discrepancies emerged when changing the A biomarker but were more obvious when changing the T biomarker. This was mainly because of the low agreement between t-tau and p-tau, regardless the high correlation between both measures. When using t-tau, the A1T1 group was markedly reduced in size (252%) in favor of the A1T2 (1490%) and A2T2 (198%) groups. Thus, the proportion of subjects in the higher risk category (A1T1) was doubled when using p-tau (n 5 158, 50.3%) than when using t-tau (n 5 75, 24.5%). This effect has already been shown in healthy control classification in ADNI [9]. Despite the fact that p-tau is often proposed as more specific for AD [4], a higher diagnostic accuracy has been reported for t-tau than for p-tau in large pathologyproven cohorts (AUC 5 0.831 and 0.753 for t-tau and p-tau, respectively) [7] and in large meta-analysis [10]. Q3 It should be noted that t-tau and p-tau values are tightly correlated with each other in the context of AD [11], and that both biomarkers show a good correlation with neurofibrillary tangle load in pathologically proven cohorts [12]. In fact, t-tau has showed even higher correlation rates with the pathologic neurofibrillary load than p-tau (r 5 0.550 and 0.466 for t-tau and p-tau, respectively)
1552-5260/Ó 2017 the Alzheimer’s Association. Published by Elsevier Inc. All rights reserved.
COR 5.4.0 DTD JALZ2404_proof 8 July 2017 7:33 pm ce
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 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
110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170
I. Illan-Gala et al. / Alzheimer’s & Dementia - (2017) 1-3
web 4C=FPO
2
Fig. 1. Subject classification into the different categories and their respective progression rates from aMCI to AD dementia for more than 2 years in the ADNI cohort when (A) using florbetapir and p-tau levels (n 5 314), (B) using CSF Ab1–42 and CSF p-tau levels (n 5 314), and (C) using CSF Ab1–42 and CSF t-tau levels (n 5 306). All groups were compared with the A2T2 group, which served as the reference group. *P ,.05 (chi-square test); tau, P ,.1 (chi-square test). Q5 Abbreviations: aMCI, amnestic mild cognitive impairment; A, amyloid; AD, Alzheimer’s disease; ADNI, Alzheimer’s Disease Neuroimaging Initiative; Ab1–42, b-amyloid1–42; CSF, cerebrospinal fluid; FBP, [18F]florbetapir; p-tau, p181-tau; T, tau.
[12]. T-tau levels can be found increased in other neurogenerative and nonneurodegenerative conditions, limiting its specificity. However, p-tau levels have also been described to be altered in other neurodegenerative
and nonneurodegenerative conditions (i.e., amyotrophic lateral sclerosis or herpetic encephalitis) [11,13]. In conclusion, our results support the synergistic effect of A and T on progression to AD dementia in aMCI, but they
COR 5.4.0 DTD JALZ2404_proof 8 July 2017 7:33 pm ce
171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231
I. Illan-Gala et al. / Alzheimer’s & Dementia - (2017) 1-3
232 also highlight the importance of the selection of biomarkers 233 and the need for precise cutoff definitions to build operative 234 and robust classification schemes. Researchers and clinicians 235 should be extremely cautious when interpreting studies to es236 timate the individual risk for progression in their cohorts. 237 This might be especially important if they use a different 238 biomarker modality to assess A or T (or neurodegeneration) 239 or if they use different thresholds. If these classifications are 240 not robust, we might ascribe incorrect progression risks to a 241 given individual and find different prevalence for each 242 category among cohorts, which has important implications 243 244 both in clinical practice and clinical trials. 245 246 Acknowledgments 247 Funding: This work was supported by research grants from 248 249 the Carlos III Institute of Health, Spain (grants PI11/02425 250 and PI14/01126 to J.F., grants PI10/1878 and PI13/01532 251 to R.B., and grants PI11/03035 and PI14/1561 to A.L.) 252 jointly funded by Fondo Europeo de Desarrollo Regional 253 (FEDER), Uni on Europea, “Una manera de hacer Europa,” 254 and CIBERNED (Program 1, Alzheimer’s Disease and other 255 dementias to A.L.). This work has also been supported by a 256 “Marat o TV3” grant (20141210 to J.F.) and by Generalitat 257 de Catalunya (2014SGR-0235). I.I.-G. is supported by the 258 i-PFIS grant (IF15/00060) from the FIS, Instituto de Salud 259 Carlos III. 260 261 Disclosure: The authors report no disclosures relevant to this 262 manuscript. 263 Ignacio Illan-Gala 264Q6 Eduard Vilaplana 265 Jordi Pegueroles 266 Victor Montal 267 Daniel Alcolea 268 269 Rafael Blesa 270 Alberto Lleo 271 Memory Unit 272 Department of Neurology 273 Hospital de la Santa Creu i Sant Pau 274 Biomedical Research Institute Sant Pau 275 Universitat Aut onoma de Barcelona 276 Barcelona, Spain Q1 277 Centro de Investigaci o n Biom e dica en Red de Enfermedades 278 Neurodegenerativas, CIBERNED, Spain 279 Q2 280 281 Juan Fortea* 282 Memory Unit 283 Department of Neurology 284 Hospital de la Santa Creu i Sant Pau 285 Biomedical Research Institute Sant Pau 286 287 288 289 290 291 292
3
Universitat Autonoma de Barcelona Barcelona, Spain Centro de Investigacion Biomedica en Red de Enfermedades Neurodegenerativas CIBERNED, Spain Barcelona Down Medical Center, Fundacio Catalana de Sındrome de Down, Barcelona, Spain For the Alzheimer’s Disease Neuroimaging Initiative1
References [1] Pascoal TA, Mathotaarachchi S, Shin M, Benedet AL, Mohades S, Wang S, et al. Synergistic interaction between amyloid and tau predicts the progression to dementia. Alzheimers Dement 2017;13:644–53. [2] Desikan RS, McEvoy LK, Thompson WK, Holland D, Roddey JC, Blennow K, et al. Amyloid-b associated volume loss occurs only in the presence of phospho-tau. Ann Neurol 2011;70:657–61. [3] Fortea J, Vilaplana E, Alcolea D, Carmona-Iragui M, SanchezSaudinos MB, Sala I, et al. Cerebrospinal fluid b-amyloid and phospho-tau biomarker interactions affecting brain structure in preclinical Alzheimer disease. Ann Neurol 2014;76:223–30. [4] Jack CR, Bennett DA, Blennow K, Carrillo MC. A/T/N: an unbiased descriptive classification scheme for Alzheimer disease biomarkers. Neurology 2016;87:539–47. [5] Mattsson N, Insel PS, Donohue M, Landau S, Jagust WJ, Shaw LM, et al. Independent information from cerebrospinal fluid amyloid-b and florbetapir imaging in Alzheimer’s disease. Brain 2015; 138:772–83. [6] Jack CR, Wiste HJ, Weigand SD, Knopman DS, Mielke MM, Vemuri P, et al. Different definitions of neurodegeneration produce similar amyloid/neurodegeneration biomarker group findings. Brain 2015;138:3747–59. [7] Shaw LM, Vanderstichele H, Knapik-Czajka M, Clark CM, Aisen PS, Petersen RC, et al. Cerebrospinal fluid biomarker signature in Alzheimer’s disease neuroimaging initiative subjects. Ann Neurol 2009; 65:403–13. [8] Vos SJ, Verhey F, Fr€olich L, Kornhuber J, Wiltfang J, Maier W, et al. Prevalence and prognosis of Alzheimer’s disease at the mild cognitive impairment stage. Brain 2015;138:1327–38. [9] Pegueroles J, Vilaplana E, Montal V, Sampedro F, Alcolea D, Carmona-Iragui M, et al. Longitudinal brain structural changes in preclinical Alzheimer disease. Alzheimer Dement 2017;13:499–509. € [10] Olsson B, Lautner R, Andreasson U, Ohrfelt A. CSF and blood biomarkers for the diagnosis of Alzheimer’s disease: a systematic review and meta-analysis. Lancet 2016;15:673–84. [11] Zetterberg H. Review: tau in biofluids—relation to pathology, imaging and clinical features. Neuropathol Appl Neurobiol 2017;43:194–9. [12] Tapiola T, Alafuzoff I, Herukka SK, Parkkinen L, Hartikainen P, Soininen H, et al. Cerebrospinal fluid b-amyloid 42 and tau proteins as biomarkers of Alzheimer-type pathologic changes in the brain. Arch Neurol 2009;66:382–9. [13] Grossman M, Elman L, McCluskey L, McMillan CT, Boller A, Powers J, et al. Phosphorylated tau as a candidate biomarker for amyotrophic lateral sclerosis. JAMA Neurol 2014;71:442–7.
http://dx.doi.org/10.1016/j.jalz.2017.06.002
COR 5.4.0 DTD JALZ2404_proof 8 July 2017 7:33 pm ce
Q4
293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54
Letter The pitfalls of biomarker-based classification schemes We read with great interest the elegant work by Pascoal et al. [1] describing a synergistic effect between florbetapir uptake values and cerebrospinal fluid (CSF) p-tau levels on the 2-year risk of progression to Alzheimer’s disease (AD) dementia in patients with amnestic mild cognitive impairment (aMCI) from the Alzheimer Disease Neuroimaging Initiative (ADNI). This synergistic toxic effect between amyloid (A) and tau (T) has strong biological evidence and has been shown to accelerate cognitive decline and brain atrophy in healthy elderly [2,3]. The authors used two different biomarker modalities for subject classification. The use of CSF biomarker levels alone would have obvious practical advantages, but might influence the results. Classification schemes using AD biomarkers are highly dependent on the standardization and reproducibility of biomarker measures [4]. A and T can be measured through different biomarkers with similar accuracy [5]. These different biomarkers often correlate with each other, but when classifying a subject into normal or abnormal for a given modality, they sometimes have poor agreement [6]. This has been described as a limitation of classification schemes [4]. To address the effect of the classification method on clinical outcomes in the study of Pascoal et al., we compared the use of CSF Ab1–42 levels instead of florbetapir uptake values and the use of CSF t-tau instead of CSF p-tau values in the same group of aMCI subjects from ADNI. We first assessed the correlation between those biomarkers and their agreement for A and T classification. We then compared the prevalence for each category (A2T2, A1T2, A2T1, and A1T1) and their respective risk of progression to AD dementia in the same group of aMCI. We used the florbetapir uptake values and CSF biomarker levels provided by the
1 Data used in the preparation of this article were obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database (adni.loni. usc.edu). As such, the investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report. A complete listing of ADNI investigators can be found at http://adni.loni.usc.edu/wp-content/ uploads/how_to_apply/ADNI_Acknowledgement_List.pdf. *Corresponding author. Tel.: 134-935565986; Fax:134-935565602. E-mail address: [email protected]
ADNI database and we applied the validated thresholds for each biomarker [7]. Both CSF Ab1–42 levels and the florbetapir uptake values and CSF p-tau and t-tau levels were similarly correlated (r 5 20.695 and 0.636, for A and T, respectively). However, the concordance for subject classification was much lower for T (k 5 0.713 and 0.228 for A and T, respectively). Fig. 1 summarizes the risk of progression to AD dementia and the prevalence of each category when using (1) [18F]florbetapir and p-ptau, (2) CSF Ab1–42 and p-tau, and (3) CSF Ab1–42 and t-tau. In all cases, we replicated the synergistic effects between A and T. However, the group of isolated amyloid pathology (A1T2) emerged as an intermediate risk group (progression rates of 5%–7.1% for the A1T2 group vs. 0% to 1.6% for the A2T2 group at 2 years). These results are in agreement with the study from Vos et al. [8] that described an intermediate progression risk for the isolated amyloid pathology group in a large multicentric MCI cohort after a follow up of 2.7 years. The most important difference was, nonetheless, the dramatic change in the prevalence of the different categories. These discrepancies emerged when changing the A biomarker but were more obvious when changing the T biomarker. This was mainly because of the low agreement between t-tau and p-tau, regardless the high correlation between both measures. When using t-tau, the A1T1 group was markedly reduced in size (252%) in favor of the A1T2 (1490%) and A2T2 (198%) groups. Thus, the proportion of subjects in the higher risk category (A1T1) was doubled when using p-tau (n 5 158, 50.3%) than when using t-tau (n 5 75, 24.5%). This effect has already been shown in healthy control classification in ADNI [9]. Despite the fact that p-tau is often proposed as more specific for AD [4], a higher diagnostic accuracy has been reported for t-tau than for p-tau in large pathologyproven cohorts (AUC 5 0.831 and 0.753 for t-tau and p-tau, respectively) [7] and in large meta-analysis [10]. Q3 It should be noted that t-tau and p-tau values are tightly correlated with each other in the context of AD [11], and that both biomarkers show a good correlation with neurofibrillary tangle load in pathologically proven cohorts [12]. In fact, t-tau has showed even higher correlation rates with the pathologic neurofibrillary load than p-tau (r 5 0.550 and 0.466 for t-tau and p-tau, respectively)
1552-5260/Ó 2017 the Alzheimer’s Association. Published by Elsevier Inc. All rights reserved.
COR 5.4.0 DTD JALZ2404_proof 8 July 2017 7:33 pm ce
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 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
110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170
I. Illan-Gala et al. / Alzheimer’s & Dementia - (2017) 1-3
web 4C=FPO
2
Fig. 1. Subject classification into the different categories and their respective progression rates from aMCI to AD dementia for more than 2 years in the ADNI cohort when (A) using florbetapir and p-tau levels (n 5 314), (B) using CSF Ab1–42 and CSF p-tau levels (n 5 314), and (C) using CSF Ab1–42 and CSF t-tau levels (n 5 306). All groups were compared with the A2T2 group, which served as the reference group. *P ,.05 (chi-square test); tau, P ,.1 (chi-square test). Q5 Abbreviations: aMCI, amnestic mild cognitive impairment; A, amyloid; AD, Alzheimer’s disease; ADNI, Alzheimer’s Disease Neuroimaging Initiative; Ab1–42, b-amyloid1–42; CSF, cerebrospinal fluid; FBP, [18F]florbetapir; p-tau, p181-tau; T, tau.
[12]. T-tau levels can be found increased in other neurogenerative and nonneurodegenerative conditions, limiting its specificity. However, p-tau levels have also been described to be altered in other neurodegenerative
and nonneurodegenerative conditions (i.e., amyotrophic lateral sclerosis or herpetic encephalitis) [11,13]. In conclusion, our results support the synergistic effect of A and T on progression to AD dementia in aMCI, but they
COR 5.4.0 DTD JALZ2404_proof 8 July 2017 7:33 pm ce
171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231
I. Illan-Gala et al. / Alzheimer’s & Dementia - (2017) 1-3
232 also highlight the importance of the selection of biomarkers 233 and the need for precise cutoff definitions to build operative 234 and robust classification schemes. Researchers and clinicians 235 should be extremely cautious when interpreting studies to es236 timate the individual risk for progression in their cohorts. 237 This might be especially important if they use a different 238 biomarker modality to assess A or T (or neurodegeneration) 239 or if they use different thresholds. If these classifications are 240 not robust, we might ascribe incorrect progression risks to a 241 given individual and find different prevalence for each 242 category among cohorts, which has important implications 243 244 both in clinical practice and clinical trials. 245 246 Acknowledgments 247 Funding: This work was supported by research grants from 248 249 the Carlos III Institute of Health, Spain (grants PI11/02425 250 and PI14/01126 to J.F., grants PI10/1878 and PI13/01532 251 to R.B., and grants PI11/03035 and PI14/1561 to A.L.) 252 jointly funded by Fondo Europeo de Desarrollo Regional 253 (FEDER), Uni on Europea, “Una manera de hacer Europa,” 254 and CIBERNED (Program 1, Alzheimer’s Disease and other 255 dementias to A.L.). This work has also been supported by a 256 “Marat o TV3” grant (20141210 to J.F.) and by Generalitat 257 de Catalunya (2014SGR-0235). I.I.-G. is supported by the 258 i-PFIS grant (IF15/00060) from the FIS, Instituto de Salud 259 Carlos III. 260 261 Disclosure: The authors report no disclosures relevant to this 262 manuscript. 263 Ignacio Illan-Gala 264Q6 Eduard Vilaplana 265 Jordi Pegueroles 266 Victor Montal 267 Daniel Alcolea 268 269 Rafael Blesa 270 Alberto Lleo 271 Memory Unit 272 Department of Neurology 273 Hospital de la Santa Creu i Sant Pau 274 Biomedical Research Institute Sant Pau 275 Universitat Aut onoma de Barcelona 276 Barcelona, Spain Q1 277 Centro de Investigaci o n Biom e dica en Red de Enfermedades 278 Neurodegenerativas, CIBERNED, Spain 279 Q2 280 281 Juan Fortea* 282 Memory Unit 283 Department of Neurology 284 Hospital de la Santa Creu i Sant Pau 285 Biomedical Research Institute Sant Pau 286 287 288 289 290 291 292
3
Universitat Autonoma de Barcelona Barcelona, Spain Centro de Investigacion Biomedica en Red de Enfermedades Neurodegenerativas CIBERNED, Spain Barcelona Down Medical Center, Fundacio Catalana de Sındrome de Down, Barcelona, Spain For the Alzheimer’s Disease Neuroimaging Initiative1
References [1] Pascoal TA, Mathotaarachchi S, Shin M, Benedet AL, Mohades S, Wang S, et al. Synergistic interaction between amyloid and tau predicts the progression to dementia. Alzheimers Dement 2017;13:644–53. [2] Desikan RS, McEvoy LK, Thompson WK, Holland D, Roddey JC, Blennow K, et al. Amyloid-b associated volume loss occurs only in the presence of phospho-tau. Ann Neurol 2011;70:657–61. [3] Fortea J, Vilaplana E, Alcolea D, Carmona-Iragui M, SanchezSaudinos MB, Sala I, et al. Cerebrospinal fluid b-amyloid and phospho-tau biomarker interactions affecting brain structure in preclinical Alzheimer disease. Ann Neurol 2014;76:223–30. [4] Jack CR, Bennett DA, Blennow K, Carrillo MC. A/T/N: an unbiased descriptive classification scheme for Alzheimer disease biomarkers. Neurology 2016;87:539–47. [5] Mattsson N, Insel PS, Donohue M, Landau S, Jagust WJ, Shaw LM, et al. Independent information from cerebrospinal fluid amyloid-b and florbetapir imaging in Alzheimer’s disease. Brain 2015; 138:772–83. [6] Jack CR, Wiste HJ, Weigand SD, Knopman DS, Mielke MM, Vemuri P, et al. Different definitions of neurodegeneration produce similar amyloid/neurodegeneration biomarker group findings. Brain 2015;138:3747–59. [7] Shaw LM, Vanderstichele H, Knapik-Czajka M, Clark CM, Aisen PS, Petersen RC, et al. Cerebrospinal fluid biomarker signature in Alzheimer’s disease neuroimaging initiative subjects. Ann Neurol 2009; 65:403–13. [8] Vos SJ, Verhey F, Fr€olich L, Kornhuber J, Wiltfang J, Maier W, et al. Prevalence and prognosis of Alzheimer’s disease at the mild cognitive impairment stage. Brain 2015;138:1327–38. [9] Pegueroles J, Vilaplana E, Montal V, Sampedro F, Alcolea D, Carmona-Iragui M, et al. Longitudinal brain structural changes in preclinical Alzheimer disease. Alzheimer Dement 2017;13:499–509. € [10] Olsson B, Lautner R, Andreasson U, Ohrfelt A. CSF and blood biomarkers for the diagnosis of Alzheimer’s disease: a systematic review and meta-analysis. Lancet 2016;15:673–84. [11] Zetterberg H. Review: tau in biofluids—relation to pathology, imaging and clinical features. Neuropathol Appl Neurobiol 2017;43:194–9. [12] Tapiola T, Alafuzoff I, Herukka SK, Parkkinen L, Hartikainen P, Soininen H, et al. Cerebrospinal fluid b-amyloid 42 and tau proteins as biomarkers of Alzheimer-type pathologic changes in the brain. Arch Neurol 2009;66:382–9. [13] Grossman M, Elman L, McCluskey L, McMillan CT, Boller A, Powers J, et al. Phosphorylated tau as a candidate biomarker for amyotrophic lateral sclerosis. JAMA Neurol 2014;71:442–7.
http://dx.doi.org/10.1016/j.jalz.2017.06.002
COR 5.4.0 DTD JALZ2404_proof 8 July 2017 7:33 pm ce
Q4
293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353