Primary Progressive Aphasia
A 25-year Retrospective
M.-Marsel Mesulam, MD
Abstract: The diagnosis of primary progressive aphasia (PPA) is
made in any patient in whom a language impairment (aphasia),
caused by a neurodegenerative disease (progressive), constitutes
the most salient aspect of the clinical picture (primary). The
language impairment can be fluent or nonfluent and may or may
not interfere with word comprehension. Memory for recent
events is relatively preserved although memory scores obtained
in verbally mediated tests may be abnormal. Lesser changes in
behavior and object recognition may be present but are not the
leading factors that bring the patient to medical attention. This
selective clinical pattern is most conspicuous in the initial stages
of the disease. Progressive nonfluent aphasia and some types of
semantic dementia can be considered subtypes of PPA. Initially
brought to the attention of contemporary literature 25 years
ago, PPA has recently witnessed substantial progress related to
its neurolinguistic features, neuroanatomy, imaging, neuro-
pathology, genetics, and risk factors.
Key Words: dementia, language
(Alzheimer Dis Assoc Disord 2007;21:S8–S11)
primary progressive aphasia (PPA), I felt this would be a
proper venue for a brief retrospective account of my
experience with this syndrome.
The story can be traced back to 1975, when the
Harvard Neurological Unit, chaired by Norman Gesch-
wind, moved from Boston City Hospital to the Beth Israel
Hospital. As I had just finished my residency at Boston
City Hospital, Norman Geschwind asked me to move
with him and establish a Behavioral Neurology Unit
(BNU) at the new site.
The BNU attracted patients with disorders covering
almost all aspects of behavioral neurology. We became
particularly familiar with the progressive amnestic de-
mentia of Alzheimer disease (AD) and the aphasic
syndromes of focal strokes.
hen Bruce Miller, the guest editor for this issue,
invited me to contribute a paper on the history of
Not all cases we encountered fit these familiar
patterns. I was particularly puzzled by a subset of aphasic
patients who displayed atypical features. When asked
about her chief complaint, one such patient said: ‘‘Syntax
errors and no articlesyWords in the my head and cut
upyWriting syntax errors. Edit my workycomputer.’’ I
must admit that my first impulse was to consider her an
impostor sent to test the limits of my gullibility. Even for
Broca aphasia, the agrammatism appeared overdone. If
this was Broca aphasia, why was she not dysarthric or
hemiparetic? Even more vexing was the absence of visible
cerebrovascular lesions in Broca’s area or anywhere in the
left hemisphere. I was also intrigued by the history of
gradual progression. We were familiar with indolent
progressions of cognitive deficits, but such cases were
usually attributed to AD and tended to be associated with
memory loss, a feature that was definitely absent in this
We soon gathered a handful of such patients: they
had an isolated aphasia but no stroke; a progressive
course but no amnesia. We were puzzled enough to
proceed with a cortical biopsy in one of our patients. As
the biopsy gave no indication of AD, it seemed that we
had stumbled onto something that was neither in
contemporary textbooks of neurology nor part of the
clinical teaching at that time. Because those were the days
when one was still rewarded for unearthing precedent, I
spent the next several weeks at the basement of Countway
Library at Harvard Medical School.
I first focused on an 1892 report by Pick,1entitled
‘‘Ueber die Beziehungen der senilen Hirnatrophie zur
Aphasie,’’ in which he describes a patient he had first
encountered on November 11, 1891. The history revealed
a progressive aphasia. I first thought that this would be
the archetypal case for the cluster of unusual patients I
had seen. However, Pick also noted that the patient had
‘‘3 Jahren progressive Geda ¨ chtnisschwa ¨ he’’ (Progressive
weakness of memory for 3 years) and that he ‘‘seine Frau
mit dem Messer bedroht habe’’ (had threatened his wife
with a knife). So I had to conclude that this patient could
not be considered a perfect prototype for the relatively
pure aphasias I had been seeing.
As I continued my library research, I found a paper
by Paul Se ´ rieux in which he describes a woman, brought
to the hospital on March 11, 1891, who presented with a
Copyrightr2007 by Lippincott Williams & Wilkins
From the Cognitive Neurology and Alzheimer’s Disease Center,
Feinberg School of Medicine, Northwestern University, Chicago, IL.
Grant Support: DC008552 from the National Institute on Deafness and
other Communication Disorders and P30 AG013854 from the
National Institute on Aging.
Reprints: M.-Marsel Mesulam, MD, Cognitive Neurology and Alzhei-
mer’s Disease Center, Feinberg School of Medicine, Northwestern
University, Chicago, IL (e-mail: email@example.com).
Alzheimer Dis Assoc Disord?Volume 21, Number 4, October–December 2007
progressive loss of word comprehension and in whom ‘‘la
me ´ moire et l’intelligence de la malade e ´ taient suffisam-
ment conserve ´ es’’ (The patient’s memory and intelligence
were sufficiently preserved.). The patient died in 1897.2
The brain, examined by Dejerine3at the Salpe ˆ trie ` re,
showed bitemporal cortical atrophy and neuronal loss.
There was obviously no way of knowing whether the
patient had AD because Alzheimer would not be
reporting the pathologic pattern that now bears his name
for another 10 years, in 1906.4
I had a chance to discuss this case with Andre Roch
Lecours at a time when we were serving as representatives
of our respective countries, Canada for Andre and United
States for me, at a meeting of the Human Frontiers for
Research Organization in Strasbourg. I mentioned to
Andre the uncertain neuropathologic status of the
Dejerine/Se ´ rieux case. A few months later, Andre visited
Boston, bearing a totally unexpected gift. Through some
stratagem known only to him, he had managed to
‘‘borrow’’ the Dejerine slides. As I did not have the heart
to violate the sanctity of the material touched by
Dejerine, I did not remove the coverslips for restaining
with more up to date methods, but I did examine the cell
and myelin preparations as carefully as I could. I found
no evidence of either senile plaques or neurofibrillary
tangles. I have since considered the Dejerine/Se ´ rieux
patient the closest prototypical example I have been able
to find of the syndrome now known as PPA.
THE SLOWLY PROGRESSIVE APHASIA STAGE
Armed with the authority of a precedent and faced
with additional patients, I felt comfortable reporting an
initial set of 6 cases in 1982 under the rubric of ‘‘Slowly
Progressive Aphasia Without Generalized Dementia.’’
The term ‘‘progressive’’ was used to differentiate the
patients from stroke-caused aphasia and the word
‘‘slowly’’ was added to differentiate them from the
progressive but relatively faster course of neoplasm.5
The phrase ‘‘without generalized dementia’’ was included
to highlight the difference from typical forms of AD.
The language disorders were heterogeneous. Some
patients were fluent others not, some had word compre-
hension impairments but not others. Furthermore, the
aphasias rarely fit the canonical patterns established in
stroke. Dysarthria, an almost invariable component of
agrammatic dysfluent aphasias due to stroke, was rarely
present. Comprehension deficits were generally confined
to single words and rarely displayed the severity seen in
typical Wernicke aphasia. In contrast to patients with
stroke-induced loss of fluency, many of our patients were
intermittently dysfluent. They produced fluent speech
when allowed to engage in small talk and circum-
locutions, but became nonfluent because of word-finding
hesitations when forced to be precise. I coined the
neologism ‘‘logopenia’’ to describe this state of fluctuat-
ing fluency in patients without the frank agrammatism of
classic nonfluent aphasias.5The intermittent interruption
of fluency also showed that the fluent-nonfluent distinc-
tion, a gold standard for the classification of stroke-based
aphasias, would be of more limited usefulness in the
classification of these patients.
THE PPA DESIGNATION
Following the 1982 paper, dozens of patients were
reported in clinics around the world and I soon realized
that the terminology needed modification. First, the
phrase ‘‘slowly progressive’’ seemed redundant. Second,
the qualification of ‘‘without generalized dementia’’
seemed to be a bit of an overkill because it was becoming
quite clear that memory loss is not the only or even
necessary criterion for dementia. A patient with a
progressive aphasia could be said to have a language-
based dementia, just as patients with typical AD are said
to have a memory-based dementia.
In 1987, in an Annals of Neurology editorial Art
Asbury had asked me to write as a companion piece to the
pioneering neuropathologic study on progressive aphasia
by Kirschner and colleagues,6I proposed a change of
terminology to PPA.7
As PPA is caused by a progressive neurodegenera-
tion that goes on to invade most of the cerebral cortex,
patients eventually displayed many additional deficits. To
capture the pivotal nature of the language impairment,
Sandra Weintraub8,9and I introduced the 2-year diag-
nostic criterion according to which the aphasia had to be
the most salient deficit and the major cause of impaired
daily living activities for approximately 2 years. This
‘‘rule’’ allowed us to weed out patients with Creutzfeldt-
Jacob disease with an aphasic onset but rapid deteriora-
tion and also typical amnestic AD patients who might
eventually develop a language disorder. Exact timing of
onset in progressive disease is notoriously difficult, and we
stressed that the ‘‘2 year’’ rule was meant to be interpreted
with considerable latitude.9
These criteria were eventually codified and the PPA
diagnosis can now be made in any patient who has a
fluent or nonfluent language disorder (aphasia) that is due
to a neurodegenerative disease (progressive) and in whom
the aphasia is initially the most salient feature of the
clinical picture (primary).10,11The aphasia can interfere
with word-finding, object naming, syntax, phonology,
morphology, spelling or word comprehension. The
progression occurs in the course of years rather than
months, and the primary nature of the aphasia is
demonstrated by showing that memory for recent events,
the recognition of familiar faces and objects, reasoning,
and basic aspects of comportment are relatively preserved
at the initial stages.12The fact that the language disorder
in PPA could be fluent or nonfluent and that it could be
associated with normal or impaired word comprehension
was emphasized in the initial description and subsequent
reviews.5,8These features are now incorporated in the
Uniform Data Set used by the Alzheimer’s Disease
Centers of the United States.13
Alzheimer Dis Assoc Disord?Volume 21, Number 4, October–December 2007Primary Progressive Aphasia
r2007 Lippincott Williams & Wilkins
CONGENERS, VARIANTS, BOUNDARIES
(PNFA), semantic dementia (SD), aphasic variant of
frontotemporal dementia (FTD), temporal variant of
frontotemporal lobar degeneration (FTLD), Gogi apha-
sia, and progressive aphasia have been used, mostly
implicitly, to denote variants of PPA.14,15We prefer
the PPA term as a root diagnosis for 2 reasons: not all
progressive aphasias fulfill the PPA criteria and not all
PPA cases are caused by FTLD pathology.
The PNFA criteria of Neary et al15fit the PPA
criteria perfectly and this syndrome can be considered a
PPA subtype. The roots of the SD syndrome can be
traced to a 1975 report by Warrington.17If the original
Neary et al criteria for SD (ie, the requirement to have
both poor word comprehension and also an associative
agnosia) are interpreted literally, SD does not seem to fit
the PPA criteria. However, the Cambridge group seems to
have redefined SD as a language disorder accompanied by
lesser and variable agnosic deficits, mostly for unfamiliar
objects.16This latter definition of SD makes it a subtype
The PNFA and SD subtypes do not account for all
of PPA. The logopenia term of the 1982 paper described a
clinical state intermediate between nonfluent and fluent
aphasia.5The most comprehensive recent development to
take this ‘‘third’’ form of PPA into account is the work of
Gorno-Tempini and colleagues18who codified a ‘‘logo-
penic’’ variant of PPA. The defining criteria for these 3
PPA variants were reviewed by an international group of
investigators who met just after the September 2006
International Frontotemporal Dementia conference in
On the basis of the initial proceedings of this meeting,
convened by Gorno-Tempini, Grossman, and Hillis, we are
now subdividing our cases into 3 variants: agrammatic/
dysfluent, semantic, and logopenic. In our clinical practice,
the agrammatic/dysfluent variant (also known as PNFA) is
characterized by impairments of syntax and fluency but
preserved word comprehension; the semantic variant is
characterized by poor word comprehension but preserved
syntax and fluency; and the logopenic variant is character-
ized by interruptions of fluency due to frequent word-
finding pauses but relatively intact syntax and word
comprehension. Anomia is present in all variants but may
become the principal feature of the logopenic subtype.
As the disease progresses, PPA patients may
develop memory disorders, associative agnosias, person-
ality changes (reminiscent of the behavioral variant
FTD), motor neuron disease, or asymmetric extra-
pyramidal deficits (of the type seen in corticobasal degene-
ration), underlining the lack of rigid boundaries in
neurodegenerative syndromes.19When such additional
findings emerge, we use the designation ‘‘PPA-plus.’’
CURRENT STATE OF RESEARCH
Since the introduction of PPA to the modern
literature 25 years ago, more than 700 papers on this
subject have been published. Neurolinguistic investiga-
tions have revealed the rich spectrum of language
impairments in the areas of syntax, category-specific
naming deficits, and language comprehension.20–22The
imaging has generally shown asymmetric atrophy and
hypometabolism, more prominent in the language-domi-
nant (usually left) hemisphere.23–25The atrophy tends to
be mostly in the perisylvian region in the agrammatic/
dysfluent and logopenic variants but extends into anterior
and medial temporal cortex in the semantic variant.18,26,27
The neuropathology shows subtypes of FTLD in 60% to
70% of cases and the plaques and tangles of AD in the
others. The FTLD pathology may include focal neuronal
loss, gliosis, tauopathy, ubiquinopathy with TDP-43
proteinopathy (a pattern known as FTLD-U), and
superficial vacuolation.28,29A major conundrum in this
area is to figure out how AD pathology, which is known
to cause the greatest initial neuronal loss in entorhinal
and hippocampal areas, can account for the ‘‘aphasia
without amnesia’’ pattern that identifies the initial stages
In nonfamilial cases, the agrammatic/dysfluent
variant (PNFA) seems more closely associated with
tauopathy whereas the semantic variant may be more
closely associated with FTLD-U.30Although most PPA is
sporadic, familial cases have also been described and
linked to FTLD-U pathology and mutations in the
progranulin gene.31–33In contrast to the sporadic cases,
the familial cases display an association of FTLD-U
pathology with the agrammatic/dysfluent rather than the
semantic variant of PPA.34
In some of these progranulin mutation families,
affected members display phenotypical homogeneity for
PPA,33whereas in others some members have PPA and
others the behavioral variant of FTD.34The cellular
mechanisms that make the same mutation lead to the
behavioral variant of FTD in some family members and
to the PPA phenotype in others remains mysterious.
The frequency of learning disabilities, especially of the
dyslexic type, is higher in PPA families than in controls or
typical AD, raising the possibility that the selective
vulnerability of the language network may also be
genetically determined, at least in some of the patients.35
A critical review of these recent developments in the
imaging, neuropathology, and genetics of PPA is beyond
the scope of this paper. They are mentioned to highlight
the vigor and productivity of research in this area.
The single most pressing challenge is to discover
effective treatments. The record thus far is not very
encouraging. A small but controlled trial with bromo-
criptine has been negative36; a memantine versus placebo
trial is reaching completion; our uncontrolled clinical
experience with cholinesterase inhibitors has been dis-
appointing. Some patients with PPA benefit from speech
therapy, others learn to use communication enhance-
ment devices, and others acquire rudiments of sign
MesulamAlzheimer Dis Assoc Disord?Volume 21, Number 4, October–December 2007
r 2007 Lippincott Williams & Wilkins
language. But these are temporary measures. We find that
psychosocial interventions, support groups and targeted
educational programs are necessary components of a
comprehensive approach to patients and families.37
While we await the appearance of effective thera-
pies, PPA offers a unique experiment of nature for
exploring the molecular fingerprints that make the
language network a primary disease target and for
probing the cognitive architecture of human language as
it undergoes a slow but relentless dissolution. Considering
the progress achieved during the past 25 years, it is safe to
predict that future work on PPA will yield pivotal insights
into human language, the mechanisms of neurodegenera-
tion, and the molecular underpinnings of system-based
1. Pick A. Ueber die Beziehungen der senilen Hirnatrophie zur
Aphasie. Prager Medizinische Wochenschrift. 1892;17:165–167.
2. Se ´ rieux P. Sur un cas de surdite ´ verbale pure. Revue de Medecine.
3. Dejerine J, Se ´ rieux P. Un cas de surdite ´ verbale pure termine ´ e par
aphasie sensorielle, suivie d’autopsie. Comptes Rendues des Se´ances
de la Socie´te´ de Biologie (Paris). 1897;49:1074–1077.
4. Alzheimer A. U¨ber einen eigenartigen schweren Erkrankungspro-
zess der Hirnrinde. Neurol Zbl. 1906;25:1134.
5. Mesulam MM. Slowly progressive aphasia without generalized
dementia. Ann Neurol. 1982;11:592–598.
6. Kirschner HS, Tanridag O, Thurman L, et al. Progressive aphasia
without dementia: two cases with focal spongiform degeneration.
Ann Neurol. 1987;22:527–532.
7. Mesulam MM. Primary progressive aphasia—differentiation from
Alzheimer’s disease [editorial]. Ann Neurol. 1987;22:533–534.
8. Mesulam M-M, Weintraub S. Spectrum of primary progressive
aphasia. In: Rossor MN, ed. Unusual Dementias. London: Baillie ` re
9. Weintraub S, Mesulam M-M. Four neuropsychological profiles in
dementia. In: Boller F, Grafman J, eds. Handbook of Neuropsychology.
Vol. 8. Amsterdam: Elsevier; 1993:253–281.
10. Mesulam M-M. Primary progressive aphasia. Ann Neurol. 2001;49:
11. Mesulam M-M. Primary progressive aphasia: a language-based
dementia. N Engl J Med. 2003;348:1535–1542.
12. Wicklund A, Johnson N, Weintraub N. Preservation of reasoning in
primary progressive aphasia: further differentiation from Alzhei-
mer’s disease and the behavioral presentation of frontotemporal
dementia. J Clin Exp Neuropsychol. 2004;26:347–355.
13. Morris JC, Weintraub S, Chui HC, et al. The Uniform Data Set
(UDS): clinical and cognitive variables and descriptive data from
Alzheimer Disease Centers. Alzheimer Dis Assoc Disord. 2006;20:
14. Miller BL, Boone K, Geschwind D, et al. Pick’s disease and
concepts. Neurologist. 1999;5:205–212.
15. Neary D, Snowden JS, Gustafson L, et al. Frontotemporal lobar
degeneration. A consensus on clinical diagnostic criteria. Neurology.
16. Adlam A-LR, Patterson K, Rogers TT, et al. Semantic dementia
and fluent primary progressive aphasia: two sides of the same coin?
17. Warrington EK. The selective impairment of semantic memory.
Q J Exp Psychol. 1975;27:635–657.
18. Gorno-Tempini ML, Dronkers NF, Rankin KP, et al. Cognition
and anatomy in three variants of primary progressive aphasia.
Ann Neurol. 2004;55:335–346.
19. Kertesz A, Martinez-Lage P, Davidson W, et al. The corticobasal
degeneration syndrome overlaps progressive aphasia and fronto-
temporal dementia. Neurology. 2000;55:1368–1375.
20. Hillis AE, Oh S, Ken L. Deterioration of naming nouns versus verbs
in primary progressive aphasia. Ann Neurol. 2004;55:268–275.
21. Grossman M, Price C, Moore P, et al. Frontotemporal dementia:
linguistic perspective: TFD-Pick Conference, London, Ontario;
22. Thompson CK, Ballard KJ, Tait ME, et al. Patterns of language
decline in non-fluent primary progressive aphasia. Aphasiology.
23. Chawluk JB, Mesulam MM, Hurtig H, et al. Slowly progressive
aphasia without generalized dementia: studies with positron emis-
sion tomography. Ann Neurol. 1986;19:68–74.
24. Abe K, Ukita H, Yanagihara T. Imaging in primary progressive
aphasia. Neuroradiology. 1997;39:556–559.
25. Tyrrell PJ, Warrington EK, Frackowiak RS, et al. Heterogeneity in
progressive aphasia due to focal cortical atrophy. A clinical and
PET study. Brain. 1990;113:1321–1336.
26. Sonty SP, Mesulam M-M, Thompson CK, et al. Primary
progressive aphasia: PPA and the language network. Ann Neurol.
27. Mummery CJ, Patterson K, Wise RJS, et al. Disrupted temporal
lobe connections in semantic dementia. Brain. 1999;122:61–73.
28. Neumann M, Sampathu DM, Kwong LK, et al. Ubiquitinated
TDP-43 in frontotemporal lobar degeneration and amyotrophic
lateral sclerosis. Science. 2006;314:130–133.
29. McKhann GM, Albert MS, Grossman M, et al. Clinical and
pathological diagnosis of frontotemporal dementia. Arch Neurol.
30. Knibb JA, Xuereb JH, Patterson K, et al. Clinical and pathological
characterization of progressive aphasia. Ann Neurol. 2006;59:
31. Cruts M, Gijselinck I, Van der Zee J, et al. Null mutations
in progranulin cause ubiquitin-positive frontotemporal dementia
linked to chromosome 17q21. Nature. 2006;442:916–919.
32. Baker M, Mackenzie IR, Pickering-Brown SM, et al. Mutations in
progranulin cause tau-negative frontotemporal dementia linked to
chromosome 17. Nature. 2006;442:916–919.
33. Mesulam M, Johnson N, Krefft TA, et al. Progranulin mutations in
primary progressive aphasia. Arch Neurol. 2007;64:43–47.
34. Van der Zee J, Rademakers R, Engelborghs S, et al. A Belgian
ancestral haplotype harbours a highly prevalent mutation for 17q21-
linked tau-negative FTLD. Brain. 2006;129:841–852.
35. Rogalski E, Johnson N, Weintraub S, et al. Increased frequency of
learning disability in patients with primary progressive aphasia and
their first degree relatives. Arch Neurol. In press.
36. Reed DA, Johnson NA, Thompson C, et al. A clinical trial
of bromocriptine for treatment of primary progressive aphasia.
Ann Neurol. 2004;56:750.
37. Weintraub S, Morhardt DJ. Treatment, education and resources for
non Alzheimer dementia: one size does not fit all. Alzheimer Care Q.
Alzheimer Dis Assoc Disord?Volume 21, Number 4, October–December 2007Primary Progressive Aphasia
r2007 Lippincott Williams & Wilkins