Person:
Keshavan, Matcheri

Profile Picture

Email Address

AA Acceptance Date

Birth Date

Research Projects

Organizational Units

Job Title

Last Name

Keshavan

First Name

Matcheri

Name

Keshavan, Matcheri
Author's Publications: search.filters.isAuthorOfPublication.3460d1a2-9a46-4c1f-9a3b-f179d943eba3

Search Results

Now showing 1 - 10 of 41
  • Thumbnail Image
    Publication
    N-acetylcysteine in a Double-Blind Randomized Placebo-Controlled Trial: Toward Biomarker-Guided Treatment in Early Psychosis
    (Oxford University Press, 2017) Conus, Philippe; Seidman, Larry J; Fournier, Margot; Xin, Lijing; Cleusix, Martine; Baumann, Philipp S; Ferrari, Carina; Cousins, Ann; Alameda, Luis; Gholam-Rezaee, Mehdi; Golay, Philippe; Jenni, Raoul; Woo, T -U Wilson; Keshavan, Matcheri; Eap, Chin B; Wojcik, Joanne; Cuenod, Michel; Buclin, Thierry; Gruetter, Rolf; Do, Kim Q
    Abstract Biomarker-guided treatments are needed in psychiatry, and previous data suggest oxidative stress may be a target in schizophrenia. A previous add-on trial with the antioxidant N-acetylcysteine (NAC) led to negative symptom reductions in chronic patients. We aim to study NAC’s impact on symptoms and neurocognition in early psychosis (EP) and to explore whether glutathione (GSH)/redox markers could represent valid biomarkers to guide treatment. In a double-blind, randomized, placebo-controlled trial in 63 EP patients, we assessed the effect of NAC supplementation (2700 mg/day, 6 months) on PANSS, neurocognition, and redox markers (brain GSH [GSHmPFC], blood cells GSH levels [GSHBC], GSH peroxidase activity [GPxBC]). No changes in negative or positive symptoms or functional outcome were observed with NAC, but significant improvements were found in favor of NAC on neurocognition (processing speed). NAC also led to increases of GSHmPFC by 23% (P = .005) and GSHBC by 19% (P = .05). In patients with high-baseline GPxBC compared to low-baseline GPxBC, subgroup explorations revealed a link between changes of positive symptoms and changes of redox status with NAC. In conclusion, NAC supplementation in a limited sample of EP patients did not improve negative symptoms, which were at modest baseline levels. However, NAC led to some neurocognitive improvements and an increase in brain GSH levels, indicating good target engagement. Blood GPx activity, a redox peripheral index associated with brain GSH levels, could help identify a subgroup of patients who improve their positive symptoms with NAC. Thus, future trials with antioxidants in EP should consider biomarker-guided treatment.
  • Thumbnail Image
    Publication
    9.3 PSYCHOSIS BIOTYPES VERSUS CLINICAL SYNDROMES THROUGH THE PRISM OF INTRINSIC NEURAL ACTIVITY
    (Oxford University Press, 2018) Clementz, Brett; Pearlson, Godfrey; Tamminga, Carol; Sweeney, John; Keshavan, Matcheri
    Abstract Background: Deviation in level of intrinsic neural activity (ongoing brain signals recorded with EEG/MEG) is observed in psychosis. Neurophysiological models have proposed this physiological indicator as a genetically mediated core deviation in psychosis. Translational models of intrinsic activity deviations promise to identifying multiple distinct physiological mechanisms for psychosis manifestation. Intrinsic activity deviations may masquerade as higher levels of neural response in sensory cortices, but ultimately may lead to poor signal-to-noise ratios, particularly when psychosis cases are required to identify stimulus salience. Why do we not hear more about intrinsic activity as a core biomarker for any psychosis variation? An explanation is provided by the current project. The Bipolar-Schizophrenia Network for Intermediate Phenotypes (B-SNIP) published a means for categorizing psychoses by neurobiological homology via use of multiple biomarkers (psychosis Biotypes) rather than by clinical features. B-SNIP demonstrated the superiority of Biotypes versus DSM diagnoses for capturing neurobiological similarity through multiple external validating measures (social functioning, measures of brain volume from structural magnetic resonance images, clinical diagnoses and biomarker features among first-degree relatives). Independent analyses since the initial publication have provided additional support for the usefulness of psychosis Biotypes. Methods: For this project, we analyzed ongoing neural activity from 64 EEG sensors during 150 intervals of 10 sec duration from over 1450 B-SNIP subjects. These data (never before published) were from the inter-trial interval (ITI) of an auditory paired-stimuli task used in Biotypes construction (these ITI data themselves were not used). Although the subjects were engaged in a task (counting the number of stimulus pairs), the data used here were not part of the task itself. Data were evaluated for single trial power (estimate of neural response strength on individual trials) as a function of frequency of neural oscillations (from 2–50 Hz) over the whole head. Data were then averaged over single trials to yield an estimate of the overall strength of nonspecific (unrelated to sensory processing) neural activity. Results: When evaluated by DSM diagnoses (schizophrenia, schizoaffective disorder, bipolar disorder), the 95% confidence intervals for all groups overlapped the healthy group means across all frequencies. When considered by psychosis Biotypes, differences were obvious and statistically significant. In comparison to healthy persons, Biotype-2 probands (the most neurophysiologically activated subgroup in previous analyses) were notably high on nonspecific neural activity, and Biotype-1 probands (the most cognitively and neurophysiologically compromised subgroup in previous analysis) were notably low. Group separations on this metric were better than those obtained with original intrinsic EEG measure used in psychosis Biotypes construction, indicating this more pure intrinsic activity measure is capturing a meaningful component of Biotype neurophysiology. This was true across a range of oscillatory frequencies for the probands. The first-degree relatives of the Biotype probands showed similar patterns, although higher frequency oscillations (above 20 Hz) better differentiated relatives from healthy persons. Discussion Intrinsic activity deviation is a promising biomarker for translational research programs aimed at differential treatment development, but using DSM psychosis diagnoses would obscure its importance for understanding psychosis.
  • Thumbnail Image
    Publication
    O10.5. ABNORMAL MODULAR ORGANIZATION OF THE FUNCTIONAL CONNECTOME PREDICTS CONVERSION TO PSYCHOSIS IN CLINICAL HIGH-RISK YOUTH
    (Oxford University Press, 2018) Collin, Guusje; Seidman, Larry; Keshavan, Matcheri; Qi, Zhenghan; Stone, William; Zhang, TianHong; Tang, Yingying; Shenton, Martha; Wang, Jijun; Whitfield-Gabrieli, Susan
    Abstract Background: The first episode of schizophrenia is typically preceded by a prodromal phase characterized by sub-threshold symptoms and declining functioning. Elucidating the neurobiological substrate of prodromal symptoms that progress into overt psychotic illness is crucial to the development of early detection and intervention strategies for schizophrenia. In this study, we performed a functional connectome analysis in a large group of adolescents and young adults at Clinical High Risk (CHR) for schizophrenia. We aim to assess whether, and if so how, baseline connectome organization distinguishes CHR youth that go on to develop psychosis. Methods: This study comprises a total of 251 subjects, including 158 psychotropically-naïve CHR subjects (CHRs) and 93 healthy controls (HCs), who were matched to CHRs on age, gender, and level of education. Prodromal symptoms and cognition were assessed using the SIPS structured interview and MATRICS cognitive battery. Anatomical T1 MRI and resting-state fMRI scans were collected at baseline and processed using Freesurfer v6.0 and CONN v17.d software. For each subject, a functional connectome map was reconstructed consisting of 162 nodes representing 148 cortical regions from the Destrieux atlas and 14 subcortical structures. Functional connectomes were analyzed in terms of modular topology using the Louvain community detection method. Modular network partitions of individual CHRs were compared to a group-averaged HC network using the rand similarity coefficient (SR), providing a measure of the level of (ab)normality of the CHRs’ modular partitions. Analysis of covariance (correcting for age- and gender) was used to compare SR levels between CHRs who developed psychosis during follow-up (CHR+; N = 23) as compared to CHRs who did not develop psychosis (CHR-; N = 135). Kaplan-Meier analysis was used to estimate psychosis-free survival functions for CHRs with below- versus above-average SR, which were compared using log-rank tests. Cox regression analysis was used to assess how baseline connectome organization and clinical measures (i.e., demographics, symptoms, IQ) predicted time to conversion. Results: Modular community detection in HCs yielded five major modules including a posterior ‘visual’, central ‘sensorimotor’, medial frontoparietal ‘default-mode’, lateral frontoparietal ‘central-executive’, and inferior ‘limbic’ module. Modular connectome organization of CHR+ was significantly less similar to HCs than CHR- (F(1,154) = 7.14, p = 0.008). A region-specific analysis to identify which regions contributed most to aberrant modular connectome organization in CHR+ showed that superior temporal (including STG), medial temporal (including amygdala), and ventromedial prefrontal regions were most abnormal in terms of their modular assignment. Psychosis-free survival functions of CHRs with low versus high SR were significantly different (z = 2.5, p = 0.013), with a Hazard ratio of 3.3 indicating an over 3-fold relative event rate (i.e., conversion to psychosis) in CHRs with abnormal baseline connectome organization. Cox regression analysis indicated that baseline connectome organization (z = -2.3, p = 0.019), IQ (z = -2.7, p = 0.007), and gender (z = 2.0, p = 0.048) predicted time to conversion. Discussion This study indicates that abnormalities in functional connectome organization precede the first psychotic episode. Conversion to psychosis was found to be over three times more likely in CHRs with abnormal modular organization of the functional connectome at baseline. Our results suggest that functional connectome reorganization may underlie the gradual manifestation of prodromal symptoms. These findings may contribute to early diagnosis and intervention in schizophrenia.
  • Thumbnail Image
    Publication
    T22. PITUITARY GLAND VOLUME DIFFERENCES IN INDIVIDUALS WITH PSYCHOSIS: RESULTS FROM THE BIPOLAR-SCHIZOPHRENIA NETWORK ON INTERMEDIATE PHENOTYPES (B-SNIP) STUDY
    (Oxford University Press, 2018) Guimond, Synthia; Tingue, Samantha; Devenyi, Gabriel A; Tang, Yun-Xiang; Mike, Luke; Mallar Chakravarty, M; Sweeney, John A; Pearlson, Godfrey D; Clementz, Brett A; Tamminga, Carol A; Keshavan, Matcheri
    Abstract Background: When exposed to stress, the hypothalamic-pituitary-adrenal axis is hyperactivated, which can cause the enlargement of the pituitary gland. Hence, pituitary gland volume could be a biomarker of stress present in psychosis. However, it remains unclear if individuals with psychosis have larger pituitary gland than healthy people. Previous studies investigating this question used small samples and reported inconsistent results. In the current study, we used an automated multi-atlas segmentation method to investigate the differences between pituitary gland volumes in a large sample of individuals on the psychosis spectrum. Methods: Data collection was completed across six sites in the Bipolar-Schizophrenia Network on Intermediate Phenotypes (B-SNIP) consortium with a total of 755 participants included in the study - 174 individuals with schizophrenia (SZ), 115 with schizoaffective disorder (SZA), 167 with psychotic bipolar disorder (PBD), and 299 healthy controls (HC). Structural magnetic resonance images were acquired and pituitary gland volumes were obtained using the automated MAGeT-Brain algorithm. General linear model and post-hoc independent t-tests were used to analysis the differences between subgroups of patients using clinical diagnosis and agnostic Biotype classification (Biotype 1 being the most cognitively impaired). We also explored potential effect of antipsychotic intake, symptoms severity and duration of illness. In all analyses, we used Bonferroni correction for multiple comparisons and entered confounds as covariates (age, sex, race, intracranial volume, and site). Results: Overall, the pituitary gland volumes were not significantly different between patients and HC. No significant main effect of diagnosis was observed, but SZ patients had trending larger pituitary volume compared to HC (p=.033, uncorrected). We observed a significant main effect of Biotype (p=.003), with Biotype 1 having significantly larger pituitary gland than HC and Biotype 2 (p=.004 and p=.013). In the patients group, no significant relationship between the pituitary gland and the amount of antipsychotic intake was observed (r=.02, p=.68). Significant correlations with the pituitary gland volume were observed with symptoms severity (r=.22, p=.000), and with the duration of illness (r=-.18, p=.002). Importantly, Biotypes did not significantly differ in terms of symptoms severity nor duration of illness. Discussion As a group, individuals with psychosis do not have abnormal pituitary gland volume, but larger pituitary gland is related to shorter duration of illness and greater symptoms severity. Therefore, larger pituitary gland volume could be a state-related biomarker of psychosis. Moreover, while we did not observe any significant subgroup differences using clinical diagnosis, our results suggest an increase in pituitary volume in biotype 1 patients compared to HC. These findings clarify previous inconsistent reports, and encourage further investigation of stress biomarkers in individual with psychosis with lower cognitive abilities. In the future, this could lead to the development of more targeted treatments for this specific subgroup of patients.
  • Thumbnail Image
    Publication
    O6.4. AUDITORY AND LANGUAGE AREAS DISTINGUISH CONVERTERS FROM NON–CONVERTERS AT BASELINE IN SHARP CLINICAL HIGH-RISK SUBJECTS FOR PSYCHOSIS STUDY
    (Oxford University Press, 2018) Del Re, Elisabetta; Stone, William; Bouix, Sylvain; Somes, Nathaniel; Li, Huijun; Tang, YinYin; Zhang, TianHong; Whitfield-Gabrieli, Susan; McCarley, Robert; Seidman, Larry J; Keshavan, Matcheri; Wang, JiJun; Shenton, Martha; Niznikiewicz, Margaret
    Abstract Background: Frontal and temporal lobes abnormalities are often reported in schizophrenia. In the present study, we tested whether or not these abnormalities exist in individuals at clinical high risk for psychosis (CHR), and whether they distinguish between those CHR who convert to psychosis versus those who do not convert to psychosis at one year. We analyzed both cortical thickness (CT) and surface area (SA) given the fact that CT and SA develop along different developmental genetically mediated pathways. Since CHR individuals also experience a deterioration of cognitive functions and sub-threshold psychotic symptoms, we also explored the relationship between cognition and symptomatology and the two brain regions. Methods: Magnetic resonance images, clinical and cognitive data were acquired in 130 CHR who did not convert to psychosis (CHR-NC), 22 CHR who converted to psychosis (CHR-C) and 92 healthy controls (HC) at the Shanghai Mental Health Center, in Shanghai, China, who were tested as part of a NIH funded China and Harvard Medical School collaboration. An internal pipeline developed at the Psychiatry Neuroimaging Laboratory (PNL), Brigham and Women’s Hospital, Harvard Medical School, was used to process the scans. The pipeline includes several quality control steps and FreeSurfer 5.3 (FS) processing, the latter modified to include an automated PNL developed masking methodology, the MABS. FS output was 9 temporal and 11 frontal regions in the left and right hemisphere. All data were Z-scored to the mean and standard deviation of HC. Gender and group differences were investigated using multivariate analyses, and Spearman’s correlations were employed to investigate the relationship between brain measures and cognitive and clinical measures. Results: SA analysis of the frontal and temporal lobes showed no significant differences among the three groups, while specific and significant group differences were found in CT. More specifically, for the temporal lobe a main effect of Group (p=0.021) and a significant interaction of Region x Group (p=0.01) were found. Post hoc analyses showed that CT of Heschl’s gyrus and of the posterior region of the superior temporal sulcus distinguished CHR-C from CHR-NC (p=0.027) and from NC (p=0.002), with CT of CHR
  • Thumbnail Image
    Publication
    A crossroad for validating digital tools in schizophrenia and mental health
    (Nature Publishing Group UK, 2018) Torous, John; Staples, Patrick; Barnett, Ian; Onnela, Jukka-Pekka; Keshavan, Matcheri
  • Thumbnail Image
    Publication
    Polygenic risk for schizophrenia and measured domains of cognition in individuals with psychosis and controls
    (Nature Publishing Group UK, 2018) Shafee, Rebecca; Nanda, Pranav; Padmanabhan, Jaya; Tandon, Neeraj; Alliey-Rodriguez, Ney; Kalapurakkel, Sreeja; Weiner, Daniel; Gur, Raquel E.; Keefe, Richard S. E.; Hill, Scot K.; Bishop, Jeffrey R.; Clementz, Brett A.; Tamminga, Carol A.; Gershon, Elliot S.; Pearlson, Godfrey D.; Keshavan, Matcheri; Sweeney, John A.; McCarroll, Steven; Robinson, Elise
    Psychotic disorders including schizophrenia are commonly accompanied by cognitive deficits. Recent studies have reported negative genetic correlations between schizophrenia and indicators of cognitive ability such as general intelligence and processing speed. Here we compare the effect of polygenetic risk for schizophrenia (PRSSCZ) on measures that differ in their relationships with psychosis onset: a measure of current cognitive abilities (the Brief Assessment of Cognition in Schizophrenia, BACS) that is greatly reduced in psychotic disorder patients, a measure of premorbid intelligence that is minimally affected by psychosis onset (the Wide-Range Achievement Test, WRAT); and educational attainment (EY), which covaries with both BACS and WRAT. Using genome-wide single nucleotide polymorphism (SNP) data from 314 psychotic and 423 healthy research participants in the Bipolar-Schizophrenia Network for Intermediate Phenotypes (B-SNIP) Consortium, we investigated the association of PRSSCZ with BACS, WRAT, and EY. Among apparently healthy individuals, greater genetic risk for schizophrenia (PRSSCZ) was significantly associated with lower BACS scores (r = −0.17, p = 6.6 × 10−4 at PT = 1 × 10−4), but not with WRAT or EY. Among individuals with psychosis, PRSSCZ did not associate with variations in any of these three phenotypes. We further investigated the association between PRSSCZ and WRAT in more than 4500 healthy subjects from the Philadelphia Neurodevelopmental Cohort. The association was again null (p > 0.3, N = 4511), suggesting that different cognitive phenotypes vary in their etiologic relationship with schizophrenia.
  • Thumbnail Image
    Publication
    T201. THE STUDY OF WHITE MATTER MATURATION IN THREE POPULATIONS OF GENETIC HIGH RISK FOR SCHIZOPHRENIA INDIVIDUALS SPANNING THE DEVELOPMENTAL TIMELINE
    (Oxford University Press, 2018) Lyall, Amanda; Somes, Nathaniel; Zhang, Fan; Robertson, James; O’Donnell, Lauren J; Rathi, Yogesh; Pasternak, Ofer; Savadjiev, Peter; Styner, Martin; Fitzgerald, Zachary; Mesholam-Gately, Raquelle; Thermenos, Heidi; Whitfield-Gabrieli, Susan; Keshavan, Matcheri; DeLisi, Lynn; Gilmore, John; Seidman, Larry J; Kubicki, Marek
    Abstract Background: While the etiology of schizophrenia (SZ) is still unclear, it has been characterized as a neurodevelopmental disorder because patients exhibit deviations from normal maturational trajectories that are evident prior to the onset of psychotic symptoms. White matter (WM) has been purported to play a central role in the development of SZ, however, the timing and nature of WM changes in SZ is still poorly understood. This study uses diffusion imaging from three independent Genetic High Risk (GHR) populations spanning the developmental timeline from infancy to young adulthood. The aim of this study is to understand the extent and the time-course of WM maturational pathologies as a function of age and genetic risk for psychosis. Methods: Two datasets of 3T diffusion-weighted images of children aged 7 to 12 (24 HC and 16 at GHR) and young adults aged 19 to 29 (26 HC and 43 GHR) were collected at the Massachusetts Institute of Technology. The third dataset of 3T images of infants aged 2 years (35 HC and 18 GHR) was collected at the University of North Carolina – Chapel Hill. Whole brain two-tensor tractography was performed and 4 bilateral WM tracts (arcuate fasciculus (AF); inferior longitudinal fasciculus (ILF); cingulum bundle (CB); superior longitudinal fasciculus-ii (SLF-ii)), were extracted utilizing an atlas-guided fiber clustering algorithm. The fractional anisotropy of the tissue (FA-t) was obtained. We carried out group comparisons of FA-t between GHR and HCs utilizing Mann-Whitney-U tests and Cohen’s d effect sizes for each WM tract. Results: Preliminary analyses reveal significant reductions in FAt between GHR and HC in the right CB (p = 0.013) in the child GHR population. This is mirrored by medium to large effect sizes in the bilateral CB in GHR children (CB-left, d = 0.51; CB-right, d = 0.79). Reductions in FAt in the adult GHR population within the right CB was the largest effect observed in the adult analysis (CB-right, d = 0.46). Effect sizes in the bilateral CB were minimal in the infant GHR population (CB-left, d = 0.14, CB-right, d = 0.11). Significant decreases were also seen in the right SLF-ii in the adult GHR population (p = 0.012), but not in the infant or child GHR populations, though the reductions in FAt in the child GHR population exhibited a small effect (d = 0.35). All other white matter tracts in the adult analysis showed minor effects ranging from d = 0.033 (ILF-right) to 0.28 (ILF-left). The children and infant population also exhibited small effect sizes for all other tracts, with the child GHR dataset ranging from 0.036 (ILF-left) to 0.41 (ILF-right) and the infant GHR dataset ranging from d = 0.038 (SLF-left) to 0.34 (ILF-left). Discussion Our preliminary results suggest that abnormal WM maturation may occur in the right CB and right SLF-ii in individuals with increased genetic risk for SZ, specifically after early childhood (7 to 12 years) and into adulthood (19 to 29 years). The CB and SLF-ii are highly implicated in working memory performance, an ability that retrospective studies have shown begins to decline during the peripubertal period in those that develop SZ (~7 to 9 years). The lack of structural findings in GHR infants, may suggest that WM alterations are more likely to arise later in development, thereby possibly identifying childhood as a vulnerable period. Taken together, the preliminary results of this study provide possible evidence of subtle divergences from a healthy WM maturational trajectory in the right CB and right SLF-ii in early to late childhood that may persist into adulthood and these deviations may contribute to cognitive phenotypes described in other studies.
  • Thumbnail Image
    Publication
    S105. VALIDATING THE PREDICTIVE ACCURACY OF THE NAPLS-2 PSYCHOSIS RISK CALCULATOR IN A CLINICAL HIGH-RISK SAMPLE FROM THE SHARP (SHANGHAI AT RISK FOR PSYCHOSIS) PROGRAM
    (Oxford University Press, 2018) Zhang, TianHong; Li, HuiJun; Xu, LiHua; Tang, YingYing; Cui, HuiRu; Wang, Junjie; Li, Chunbo; Woodberry, Kristen; Shapiro, Daniel I; Niznikiewicz, Margaret; Shenton, Martha; Keshavan, Matcheri; Stone, William; Wang, JiJun; McCarley, Robert W; Seidman, Larry J
    Abstract Background: The present study aims to validate the predictive accuracy of the NAPLS-2 psychosis risk calculator in a clinical high-risk (CHR) sample from the SHARP (ShangHai At Risk for Psychosis) program in Shanghai, China using comparable inclusion/exclusion criteria and assessments. Methods: Three hundred CHR individuals were identified by the Chinese version of the Structured Interview for Prodromal Symptoms. Of these, 228 (76.0%) completed neuro-cognitive assessments at baseline and 199 (66.3%) had at least a one-year follow-up assessment. The latter group was used in risk calculation. Six key predictors (baseline age, unusual thoughts and suspiciousness, symbol coding and verbal learning test performance, functional decline and family history of psychosis) were entered into the NAPLS-2 model to generate a psychosis risk estimate for each case. The area under the receiver operating characteristic curve (AUC) was used to test the effectiveness of this discrimination. Results: The NAPLS risk calculator showed moderate discrimination of subsequent transition to psychosis in the SHARP sample with an AUC of 0.631 (p = 0.007). Whether discriminating either transition or poor treatment/clinical outcomes, the AUC of the model increased to 0.754 (p < 0.001). A risk estimate of 30% or higher had moderate sensitivity (53%) and excellent specificity (86%) for prediction of poor treatment/clinical outcome. Discussion The NAPLS-2 risk calculator largely generalizes to a Shanghai CHR sample but is meaningfully improved when predicting an individual’s poor clinical outcome as well as conversion. Our findings provide a critical step in the implementation of CHR risk calculation in China.
  • Thumbnail Image
    Publication
    21.4 BASELINE CLINICAL AND BIOLOGICAL VARIABLES PREDICTING 1 YEAR OUTCOME OF SUBJECTS AT CLINICAL HIGH RISK OF PSYCHOSIS: INSIGHT FROM SHANGHAI AT RISK FOR PSYCHOSIS (SHARP) PROGRAM
    (Oxford University Press, 2018) Zhang, TianHong; Li, HuiJun; Tang, YingYing; Li, Chunbo; Woodberry, Kristen; Shapiro, Daniel I; Niznikiewicz, Margaret; Shenton, Martha; Keshavan, Matcheri; Stone, William; Wang, Jijun
    Abstract Background: In 2010, the “ShangHai At Risk for Psychosis (SHARP)” study was launched at the Shanghai Mental Health Center (SMHC), the largest outpatient mental health clinic in China. The Chinese SHARP research was led by Dr. Larry Seidman, who was also the PI of the Harvard site of the NAPLS project. He had implemented methods very similar to those used in NAPLS for the identification of clinical high risk (CHR) individuals in Mainland China in studies jointly funded by the United States National Institute of Mental Health and Chinese funding agencies. Methods: Dr. Seidman began a collaboration with the SMHC by advising us in carrying out an epidemiological study and then received joint funding for an R21 MH093294 (Fogarty/NIH, “Broadening the Investigation of Psychosis Prodrome to Different Cultural Groups”) designed to implement a variety of clinical, neurocognitive and event related potential (ERP) measures in a preliminary study of CHR. That study, which began in April 2012 and ended in March 2015, aimed to build research capacity at the SMHC. He guide us provided 4 in-person research and clinical skills trainings (2 in SMHC, 2 in Boston), translated a widely used CHR diagnostic instrument (the Structured Interview for Prodromal Symptoms/SIPS), and trained China partners to conduct a preliminary study of 100 CHR individuals. Building upon the R21 project, and the North American Prodrome Longitudinal Studies (NAPLS) model, the same group of researchers led by Dr. Seidman is collaborating on an NIMH R01 101052-01 project (2013 to 2016) to examine biomarkers of CHR with 1 year follow-up. The R21 and R01 collaborations between Harvard Medical School (HMS), MIT, Florida A&M University (FAMU), and SMHC investigators have capitalized on the resources and experiences of the Harvard researchers as members of NAPLS, MIT researchers’ leading role in functional magnetic resonance imaging (fMRI), and FAMU researchers’ expertise in bridging western and Chinese cultures to enhance the existing capacity of Chinese researchers studying the biopsychosocial aspects of CHR. Finally, a stratified cohort of 300 CHR participants was recruited between 2012–2015, and followed up for at least 1 year. Results: With the hope of Dr. Seidman, the SHARP project is ongoing and getting better, larger and stronger. Of the total 417 CHR participants (previous epidemiological survey [CHR, n = 117], R21 [CHR, n = 100], R01 [CHR, n =200]), 349 completed at least a year of follow-up (until August 30, 2017; the longest follow-up case was six and a half years), in which 83 converted to psychosis, and 68 were lost. Preliminary data showed about 20% CHR converted to a psychotic disorder over the course of follow-up, several clinical factors such as 1) functional decline; 2) selected positive symptoms(unusual thoughts and suspiciousness); 3) selected negative symptoms(social anhedonia, expression of emotion, and ideational richness); biological factors such as the P300 auditory ERP; fMRI: Reduced anti-correlation between the bilateral parietal lobule and left dorsolateral prefrontal cortex; Structural MRI: superior temporal gyrus. et al. are account for increasing the risk of conversion to full psychosis. Discussion This is the first, well-implemented, longitudinal study of CHR in a low and middle-income country to comprehensively investigate clinical and biological factors in predicting psychosis conversion and illness progression. Dr. Seidman provide a critical step in the implementation of CHR concept in China, just as an obvious need and urgency for prevention and early intervention for Chinese patients with schizophrenia.