Letters to the Editor
964 5 Anisman H, Du L, Palkovits M, Faludi G, Kovacs GG, Szontagh-Kishazi P et al. J Psychiatry Neurosci 2008; 33: 131–141. 6 Schmidt EF, Warner-Schmidt JL, Otopalik BG, Pickett SB, Greengard P, Heintz N. Cell 2012; 149: 1152–1163. 7 Egeland M, Warner-Schmidt J, Greengard P, Svenningsson P. Biol Psychiatry 2010; 67: 1048–1056.
Supplementary Information accompanies the paper on the Molecular Psychiatry website (http://www.nature.com/mp)
Dopamine transporter imaging: nonindependence of regional measures Molecular Psychiatry (2014) 19, 964; doi:10.1038/mp.2014.15; published online 4 March 2014
We read with interest the systematic review and meta-analysis by Faroane et al.1 of the association between the SLC6A3 3’UTR polymorphism and in vivo striatal dopamine transporter (DAT) binding measured using molecular imaging studies. They found that the 9R allele of dopamine active transporter (DAT) gene (SLC6A3) was associated with increased DAT activity in human adults. This conclusion was based on the single-photon emission computed tomography and positron emission tomography (PET) studies. However, we have a methodological concern that has a major implication for some of their findings. The findings from the PET imaging were based on two studies that used two different ligands1 (Figure 2). The analysis, as illustrated in the funnel plot1 (Figure 2), shows that the standardiazed mean differences from different striatal subregions within the same study were treated as independent measures. Clearly the DAT genotype is the same for a given subject irrespective of the region, and as such these are not independent measures. Furthermore, molecular imaging measures of dopamine in striatal subregions are highly correlated2 and not independent due to partial volume effects.3,4 For these two reasons the analysis appears to have violated the requirement in a meta-analysis for measures to be independent.5 Additionally, while it is possible to combine the effect sizes from meta-analysis with two or more studies, at least N = 6 has been suggested as a minimum below which findings are likely to be unreliable.6 In meta-analysis of n = 2 studies, heterogeneity analyses are likely to underestimate the variance among studies.7 Finally, the inconsistencies between results of imaging studies may also be due to biological differences in the DAT protein expression, as amine transporters are post-transcriptionally regulated by various kinases and phosphatases.8,9 For these reasons we are concerned that the conclusions may not be justified1 and welcome clarification. CONFLICT OF INTEREST
REFERENCES 1 Faraone SV, Spencer TJ, Madras BK, Zhang-James Y, Biederman J. Mol Psychiatry advance online publication, 24 September 2013; doi:10.1038/mp.2013.126 (e-pub ahead of print). 2 Egerton A, Demjaha A, McGuire P, Mehta MA, Howes OD. Neuroimage 2010; 50: 524–531. 3 Kim E, Shidahara M, Tsoumpas C, McGinnity CJ, Kwon JS, Howes OD et al. J Cereb Blood Flow Metab 2013; 33: 914–920. 4 Zald DH, Woodward ND, Cowan RL, Riccardi P, Ansari MS, Baldwin RM et al. Neuroimage 2010; 51: 53–62. 5 Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis JP et al. Br Med J 2009; 339: b2700. 6 Fu R, Gartlehner G, Grant M, Shamliyan T, Sedrakyan A, Wilt TJ et al. J Clin Epidemiol 2011; 64: 1187–1197. 7 Kontopantelis E, Springate DA, Reeves D. PLoS One 2013; 8: e69930. 8 Morón JA, Zakharova I, Ferrer JV, Merrill GA, Hope B, Lafer EM et al. J Neurosci 2003; 23: 8480–8488. 9 Ramamoorthy S, Shippenberg TS, Jayanthi LD. Pharmacol Ther 2011; 129: 220–238.
Response to Selveraj et al. Molecular Psychiatry (2014) 19, 964–965; doi:10.1038/mp.2014.26; published online 1 April 2014 Our recent meta-analysis published in Molecular Psychiatry1 analysed two positron emission tomography (PET) studies and ten single photon emission tomography (SPECT) studies. We concluded that the 9R allele of the 3’UTR polymorphism of the dopamine transporter gene (SLC6A3) regulates dopamine activity in striatal brain regions independent of the presence of neuropsychiatric illness. A Letter to the Editor by Selveraj et al.2 raised several questions about our work, which we would like to address. Our meta-analysis presented PET results from the two extant studies. The Letter is concerned with the small number of studies. We agree with this concern, which is why we addressed it as a limitation in our paper. The Letter is also concerned that we used several potentially correlated measures from the same studies. Although this issue would not bias our estimates of the standardized mean difference (SMD) effect size or the confidence intervals for each study, it could have affected our significance levels for the pooled analyses. We allay this latter concern with three observations. First, in Figure 2 from our paper readers can see that all of the 11 PET measures had SMDs >1.0, suggesting that the 9R allele is associated with increased dopamine transporter (DAT) binding. Second, we re-analysed data separately for the two brain regions that were included in both PET studies. In these pooled analyses, the SMDs are statistically significant for both caudate (SMD = 0.38, z = 2.4, P = 0.017) and putamen (SMD = 0.33, z = 2.1, P = 0.04). Third, our conclusions are consistent with the conclusions reported by each of the PET studies.3,4 Thus, we stand by our original conclusions but refer readers of this letter to our paper for an understanding of factors that limit the strength of these conclusions. Finally, the Letter also notes that ‘inconsistencies between results of imaging studies may also be due to biological differences in the DAT protein expression, as amine transporters are post-transcriptionally regulated by various kinases and phosphatases.’ We agree (as is clear in our manuscript’s Discussion) that the regulation of DAT expression is complex. We welcome further research to clarify this issue.
The authors declare no conflict of interest.
S Selvaraj1, F Turkheimer2 and O Howes1,2 1 Medical Research Council Clinical Sciences Centre, Institute of Clinical Sciences, Hammersmith Hospital Campus, Imperial College London, London, UK and 2 King’s College London, Institute of Psychiatry, London, UK E-mail: [email protected] Molecular Psychiatry (2014), 960 – 965
CONFLICT OF INTEREST The authors declare no conflict of interest.
SV Faraone1, TJ Spencer2, BK Madras3,4, Y Zhang-James1 and J Biederman2 1 Departments of Psychiatry and of Neuroscience and Physiology, SUNY Upstate Medical University, Syracuse, NY, USA; © 2014 Macmillan Publishers Limited
Letters to the Editor 2
Pediatric Psychopharmacology Unit, Psychiatry Service, Department of Psychiatry, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA; 3 New England Primate Research Center, Division of Neuroscience, Southborough, MA, USA and 4 Harvard Medical School, Boston, MA, USA E-mail: [email protected]
© 2014 Macmillan Publishers Limited
965
REFERENCES 1 Faraone SV, Spencer TJ, Madras BK, Zhang-James Y, Biederman J. Mol Psychiatry 2014; 19: 880–889. 2 Selvaraj S, Turkheimer F, Howes O. Mol Psychiatry 2014; 19: 964. 3 Shumay E, Chen J, Fowler JS, Volkow ND. PLoS ONE 2011; 6: e22754. 4 Spencer TJ, Biederman J, Madras BK, Dougherty DD, Bonab AA, Livni E et al. Biol Psychiatry 2007; 62: 1059–1061.
Molecular Psychiatry (2014), 960 – 965
964 5 Anisman H, Du L, Palkovits M, Faludi G, Kovacs GG, Szontagh-Kishazi P et al. J Psychiatry Neurosci 2008; 33: 131–141. 6 Schmidt EF, Warner-Schmidt JL, Otopalik BG, Pickett SB, Greengard P, Heintz N. Cell 2012; 149: 1152–1163. 7 Egeland M, Warner-Schmidt J, Greengard P, Svenningsson P. Biol Psychiatry 2010; 67: 1048–1056.
Supplementary Information accompanies the paper on the Molecular Psychiatry website (http://www.nature.com/mp)
Dopamine transporter imaging: nonindependence of regional measures Molecular Psychiatry (2014) 19, 964; doi:10.1038/mp.2014.15; published online 4 March 2014
We read with interest the systematic review and meta-analysis by Faroane et al.1 of the association between the SLC6A3 3’UTR polymorphism and in vivo striatal dopamine transporter (DAT) binding measured using molecular imaging studies. They found that the 9R allele of dopamine active transporter (DAT) gene (SLC6A3) was associated with increased DAT activity in human adults. This conclusion was based on the single-photon emission computed tomography and positron emission tomography (PET) studies. However, we have a methodological concern that has a major implication for some of their findings. The findings from the PET imaging were based on two studies that used two different ligands1 (Figure 2). The analysis, as illustrated in the funnel plot1 (Figure 2), shows that the standardiazed mean differences from different striatal subregions within the same study were treated as independent measures. Clearly the DAT genotype is the same for a given subject irrespective of the region, and as such these are not independent measures. Furthermore, molecular imaging measures of dopamine in striatal subregions are highly correlated2 and not independent due to partial volume effects.3,4 For these two reasons the analysis appears to have violated the requirement in a meta-analysis for measures to be independent.5 Additionally, while it is possible to combine the effect sizes from meta-analysis with two or more studies, at least N = 6 has been suggested as a minimum below which findings are likely to be unreliable.6 In meta-analysis of n = 2 studies, heterogeneity analyses are likely to underestimate the variance among studies.7 Finally, the inconsistencies between results of imaging studies may also be due to biological differences in the DAT protein expression, as amine transporters are post-transcriptionally regulated by various kinases and phosphatases.8,9 For these reasons we are concerned that the conclusions may not be justified1 and welcome clarification. CONFLICT OF INTEREST
REFERENCES 1 Faraone SV, Spencer TJ, Madras BK, Zhang-James Y, Biederman J. Mol Psychiatry advance online publication, 24 September 2013; doi:10.1038/mp.2013.126 (e-pub ahead of print). 2 Egerton A, Demjaha A, McGuire P, Mehta MA, Howes OD. Neuroimage 2010; 50: 524–531. 3 Kim E, Shidahara M, Tsoumpas C, McGinnity CJ, Kwon JS, Howes OD et al. J Cereb Blood Flow Metab 2013; 33: 914–920. 4 Zald DH, Woodward ND, Cowan RL, Riccardi P, Ansari MS, Baldwin RM et al. Neuroimage 2010; 51: 53–62. 5 Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis JP et al. Br Med J 2009; 339: b2700. 6 Fu R, Gartlehner G, Grant M, Shamliyan T, Sedrakyan A, Wilt TJ et al. J Clin Epidemiol 2011; 64: 1187–1197. 7 Kontopantelis E, Springate DA, Reeves D. PLoS One 2013; 8: e69930. 8 Morón JA, Zakharova I, Ferrer JV, Merrill GA, Hope B, Lafer EM et al. J Neurosci 2003; 23: 8480–8488. 9 Ramamoorthy S, Shippenberg TS, Jayanthi LD. Pharmacol Ther 2011; 129: 220–238.
Response to Selveraj et al. Molecular Psychiatry (2014) 19, 964–965; doi:10.1038/mp.2014.26; published online 1 April 2014 Our recent meta-analysis published in Molecular Psychiatry1 analysed two positron emission tomography (PET) studies and ten single photon emission tomography (SPECT) studies. We concluded that the 9R allele of the 3’UTR polymorphism of the dopamine transporter gene (SLC6A3) regulates dopamine activity in striatal brain regions independent of the presence of neuropsychiatric illness. A Letter to the Editor by Selveraj et al.2 raised several questions about our work, which we would like to address. Our meta-analysis presented PET results from the two extant studies. The Letter is concerned with the small number of studies. We agree with this concern, which is why we addressed it as a limitation in our paper. The Letter is also concerned that we used several potentially correlated measures from the same studies. Although this issue would not bias our estimates of the standardized mean difference (SMD) effect size or the confidence intervals for each study, it could have affected our significance levels for the pooled analyses. We allay this latter concern with three observations. First, in Figure 2 from our paper readers can see that all of the 11 PET measures had SMDs >1.0, suggesting that the 9R allele is associated with increased dopamine transporter (DAT) binding. Second, we re-analysed data separately for the two brain regions that were included in both PET studies. In these pooled analyses, the SMDs are statistically significant for both caudate (SMD = 0.38, z = 2.4, P = 0.017) and putamen (SMD = 0.33, z = 2.1, P = 0.04). Third, our conclusions are consistent with the conclusions reported by each of the PET studies.3,4 Thus, we stand by our original conclusions but refer readers of this letter to our paper for an understanding of factors that limit the strength of these conclusions. Finally, the Letter also notes that ‘inconsistencies between results of imaging studies may also be due to biological differences in the DAT protein expression, as amine transporters are post-transcriptionally regulated by various kinases and phosphatases.’ We agree (as is clear in our manuscript’s Discussion) that the regulation of DAT expression is complex. We welcome further research to clarify this issue.
The authors declare no conflict of interest.
S Selvaraj1, F Turkheimer2 and O Howes1,2 1 Medical Research Council Clinical Sciences Centre, Institute of Clinical Sciences, Hammersmith Hospital Campus, Imperial College London, London, UK and 2 King’s College London, Institute of Psychiatry, London, UK E-mail: [email protected] Molecular Psychiatry (2014), 960 – 965
CONFLICT OF INTEREST The authors declare no conflict of interest.
SV Faraone1, TJ Spencer2, BK Madras3,4, Y Zhang-James1 and J Biederman2 1 Departments of Psychiatry and of Neuroscience and Physiology, SUNY Upstate Medical University, Syracuse, NY, USA; © 2014 Macmillan Publishers Limited
Letters to the Editor 2
Pediatric Psychopharmacology Unit, Psychiatry Service, Department of Psychiatry, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA; 3 New England Primate Research Center, Division of Neuroscience, Southborough, MA, USA and 4 Harvard Medical School, Boston, MA, USA E-mail: [email protected]
© 2014 Macmillan Publishers Limited
965
REFERENCES 1 Faraone SV, Spencer TJ, Madras BK, Zhang-James Y, Biederman J. Mol Psychiatry 2014; 19: 880–889. 2 Selvaraj S, Turkheimer F, Howes O. Mol Psychiatry 2014; 19: 964. 3 Shumay E, Chen J, Fowler JS, Volkow ND. PLoS ONE 2011; 6: e22754. 4 Spencer TJ, Biederman J, Madras BK, Dougherty DD, Bonab AA, Livni E et al. Biol Psychiatry 2007; 62: 1059–1061.
Molecular Psychiatry (2014), 960 – 965
