Surgical decision making for deep brain stimulation should not be based on aggregated normative data mining

We read with interest the manuscript by Li et al. “Toward a unified connectomic target for deep brain stimulation in obsessive-compulsive disorder”, published on the preprint server bioRxiv (13 april 2019) [[18]Li N. Baldermann J.C. Kibleur A. et al.Toward a unified connectomic target for deep brain stimulation in obsessive-compulsive disorder.2019: 1-28Google Scholar]. We would like to comment on the far-reaching conclusions drawn from the use of aggregated normative and connectomic data for their impact on decision making for surgical targeting in deep brain stimulation (DBS). Furthermore, we would like to discuss the emerging use of publication of unreviewed scientific data in various open formats. In their work, Li et al. compare two different patient cohorts undergoing DBS for treatment resistant obsessive compulsive disorder (OCD) from two study centers, Grenoble (n = 14) and Cologne (n = 22) with DBS in two different brain targets, the anteromedial subthalamic nucleus (amSTN) in Grenoble and the anterior limb of the internal capsule = ALIC in Cologne. The results of the Cologne cohort have previously been published in a very similar context, somewhat surprisingly with different results [[5]Baldermann J.C. Melzer C. Zapf A. et al.Connectivity profile predictive of effective deep brain stimulation in obsessive-compulsive disorder.Biol Psychiatry. 2019; 85: 735-743Abstract Full Text Full Text PDF PubMed Scopus (95) Google Scholar,[18]Li N. Baldermann J.C. Kibleur A. et al.Toward a unified connectomic target for deep brain stimulation in obsessive-compulsive disorder.2019: 1-28Google Scholar]. Li et al. base their conclusion on the analysis of aggregated normative data (due to a lack of individual tractographic data in both cohorts), assuming by this that these patients have unaltered, identical white matter anatomy. They find a common “tract target” that jointly explains anti OCD effects in both groups, and they conclude that this new pathway might be of further use in DBS. These results are certainly relevant - albeit not new [[9]Coenen V.A. Schlaepfer T.E. Goll P. Reinacher P.C. Voderholzer U. Tebartz van Elst L. Urbach H. Freyer T. The medial forebrain bundle as a target for deep brain stimulation for obsessive-compulsive disorder.CNS Spectr. 2016; 493: 1-8Google Scholar,[19]Liebrand L.C. Caan M.W.A. Schuurman P.R. van den Munckhof P. Figee M. Denys D. van Wingen G.A. Individual white matter bundle trajectories are associated with deep brain stimulation response in obsessive-compulsive disorder.Brain Stimul. 2019; 12: 353-360Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar]. The article implies that the associative/limbic parts of the STN were targeted by DBS for OCD because of image derived circuitry and connectivity. Although the imaging data are highly interesting from a mechanistic viewpoint, the reality is that serendipity was the driver, after case reports emerged of improvement in concomitant OCD symptoms in patients who had undergone STN DBS for Parkinson's disease [[15]Fontaine D. Mattei V. Borg M. Langsdorff von D. Magnie M.-N. Chanalet S. Robert P. Paquis P. Effect of subthalamic nucleus stimulation on obsessive-compulsive disorder in a patient with Parkinson disease.J Neurosurg. 2004; 100 (Case report): 1084-1086Crossref PubMed Scopus (149) Google Scholar,[20]Mallet L. Polosan M. Jaafari N. et al.Subthalamic nucleus stimulation in severe obsessive-compulsive disorder.N Engl J Med. 2008; 359: 2121-2134Crossref PubMed Scopus (641) Google Scholar]. Li et al. state: […] “This final bundle may indeed represent a “tract-target” to treat OCD with DBS. Given this potential clinical importance, we characterized its anatomical properties using additional views relatively to anatomical landmarks that could be used during stereotactic planning [ …]”. [[18]Li N. Baldermann J.C. Kibleur A. et al.Toward a unified connectomic target for deep brain stimulation in obsessive-compulsive disorder.2019: 1-28Google Scholar] This is probably the most problematic sentence in this paper as it points to the use of normative data for DBS surgical planning. Neurosurgeons have moved away from normative atlases and embraced the evolution of structural MRI in stereotactic surgery (allowing direct targeting); more recently tractographic methods are being investigated for direct targeting [[12]Fenoy A.J. Schiess M.C. Deep brain stimulation of the dentato-rubro-thalamic tract: outcomes of direct targeting for tremor.Neuromodulation. 2017; 20: 429-436Crossref PubMed Scopus (54) Google Scholar,[23]Schlaepfer T.E. Bewernick B.H. Kayser S. Mädler B. Coenen V.A. Rapid effects of deep brain stimulation for treatment-resistant major depression.Biol Psychiatry. 2013; 73: 1204-1212Abstract Full Text Full Text PDF PubMed Scopus (357) Google Scholar]. Whatever the imaging technology, surgical targeting and planning are based on imaging from the individual who is undergoing surgery. The normative aggregated large cohort data approach recently offered by Horn et al. to the DBS field [[16]Horn A. Li N. Dembek T.A. et al.Lead-DBS v2: towards a comprehensive pipeline for deep brain stimulation imaging.Neuroimage. 2019; 184: 293-316Crossref PubMed Scopus (249) Google Scholar,[17]Horn A. Reich M. Vorwerk J. et al.Connectivity predicts deep brain stimulation outcome in Parkinson's disease.Ann Neurol. 2017; 82: 67-78Crossref PubMed Scopus (270) Google Scholar] can be very useful when exploring group data, but becomes problematic when it claims to guide DBS surgery in individual patients: the anatomy of large normative cohorts (n) are morphed into a unified space (MNI = Montreal Neurological Institute space) which thereby becomes an atlas, not better and probably even worse than the histological atlases based on the mid-commissural point, “augmented” with normative large cohort connectomic information (e.g. human connectome project [[18]Li N. Baldermann J.C. Kibleur A. et al.Toward a unified connectomic target for deep brain stimulation in obsessive-compulsive disorder.2019: 1-28Google Scholar]). It cannot support conclusions on where an effective electrode should ideally be placed in individual patients considering normal anatomical variations and potentially disease-related alterations of (white matter) anatomy. It is of note that the scientific field perceives this augmentation with normative data and its aggregation as a step towards individualized target definition for DBS in psychiatric diseases [[26]Voon V. Toward precision medicine: prediction of deep brain stimulation targets of the ventral internal capsule for obsessive-compulsive disorder.Biol Psychiatry. 2019; 85: 708-710Abstract Full Text Full Text PDF PubMed Scopus (5) Google Scholar]. In a stricter and stereotactic sense, it is not. Even papers using high quality patient individual connectome data to analyze lead location as a function of clinical outcome in the better understood neural pathways involved in tremor are more cautious when considering extrapolation of this approach prospectively in patients [[1]Akram H. Dayal V. Mahlknecht P. et al.Connectivity derived thalamic segmentation in deep brain stimulation for tremor.Neuroimage: Neuroimage Clin. 2018; 18: 130-142Crossref PubMed Scopus (86) Google Scholar]. Open source systems such as Lead-DBS [[16]Horn A. Li N. Dembek T.A. et al.Lead-DBS v2: towards a comprehensive pipeline for deep brain stimulation imaging.Neuroimage. 2019; 184: 293-316Crossref PubMed Scopus (249) Google Scholar] are not CE or FDA approved and they are not tools intended and cleared for patient treatment (including both surgical planning and neurological programming) but they can be perceived and used as such. These systems characterize as research tools and are used to aggregate data from different sources (and in part from undisclosed, often changing libraries) to analyze and draw conclusions on “optimal” electrode positions, and by this, on targeting. While this is stated in disclaimers, the message is that new generations of DBS clinician-scientists are offered the use of normative information to place electrodes and “test” new targets which they perceive as reality. Although in our opinion it can be useful to work with normative data for pure analysis, it is problematic if normative information is to be used for targeting in a real patient who is not part of the analyzed cohort (the n + 1 patient). The use of aggregated patient cohorts is an alternative [[1]Akram H. Dayal V. Mahlknecht P. et al.Connectivity derived thalamic segmentation in deep brain stimulation for tremor.Neuroimage: Neuroimage Clin. 2018; 18: 130-142Crossref PubMed Scopus (86) Google Scholar] that can inform targeting that must be based on individual imaging [[21]Riva-Posse P. Choi K.S. Holtzheimer P.E. Crowell A.L. Garlow S.J. Rajendra J.K. McIntyre C.C. Gross R.E. Mayberg H.S. A connectomic approach for subcallosal cingulate deep brain stimulation surgery: prospective targeting in treatment-resistant depression.Mol Psychiatry. 2017; 62: 10Google Scholar]. Until prospectively acquired data have demonstrated the superiority of targeting on the basis of new forms of normative data (beyond the traditional MCP (=midcommissural point) “libraries”) with regards to safety and clinical outcomes, authors should temper conclusions based on retrospective data collection with respect to the complex clinical-surgical decision process. A better example for trial design is the actual comparison of the two target regions (ALIC, amSTN) in a prospective blinded clinical study with a crossover design and analysis of individual connectomic data, published recently [[25]Tyagi H. Apergis-Schoute A.M. Akram H. et al.A randomized trial directly comparing ventral capsule and anteromedial subthalamic nucleus stimulation in obsessive-compulsive disorder: clinical and imaging evidence for dissociable effects.Biol Psychiatry. 2019; 85: 726-734Abstract Full Text Full Text PDF PubMed Scopus (85) Google Scholar]. The reality today remains, that targeting decisions are an amalgamation of direct imaging of visualized structures, implicit or explicit use of normative data, possible intraoperative testing in the awake patient, and immediate postoperative imaging for confirmation of lead location and targeting accuracy. The scientific community should also be alerted to the proliferation of preprint publishing sites. Websites that present un-reviewed data include the bioRxiv service, the “Lead-DBS” homepage (linked), Research gate and other social media, as well as open digital libraries [[16]Horn A. Li N. Dembek T.A. et al.Lead-DBS v2: towards a comprehensive pipeline for deep brain stimulation imaging.Neuroimage. 2019; 184: 293-316Crossref PubMed Scopus (249) Google Scholar] that in the future might contain normatively derived brain structures. The bioRxiv homepage states: “Before formal publication in a scholarly journal, scientific and medical articles are traditionally “peer reviewed.” […] “Readers should therefore be aware that articles on bioRxiv have not been finalized by authors, might contain errors, and report information that has not yet been accepted or endorsed in any way by the scientific or medical community.” [for full statement cf: Anonymous (2019) https://www.biorxiv.org/content/what-unrefereed-preprint, assessed 26 april 2019]. It is not always easy to differentiate a self-published and unreviewed preprint from an already reviewed and accepted manuscript preprint. However, this form of unreviewed preprint publishing obviously is in concordance with publication politics of at least some Journals. [Anonymous (2017) Preprints under review. nature communications. https://www.nature.com/articles/s41467-017-00950-5.pdf (accessed online 26 april 2019)]. Understanding the nature of preprint data publishing is important for the evaluation of scientific data and the above-mentioned disclaimer is helpful. Moreover, in this instance there is a potentially dangerous combination where unreviewed scientific data are made available to the DBS community, open for uncritical use in uncontrolled and unregulated open source planning and visualization environments [[16]Horn A. Li N. Dembek T.A. et al.Lead-DBS v2: towards a comprehensive pipeline for deep brain stimulation imaging.Neuroimage. 2019; 184: 293-316Crossref PubMed Scopus (249) Google Scholar]. Li et al. state: […] “Finally, we show that most if not all literature-defined DBS targets that were used to treat OCD in the past fall along the tract-target identified in the present study”. [[18]Li N. Baldermann J.C. Kibleur A. et al.Toward a unified connectomic target for deep brain stimulation in obsessive-compulsive disorder.2019: 1-28Google Scholar] This pathway has previously been published as a target for DBS in major depression [[23]Schlaepfer T.E. Bewernick B.H. Kayser S. Mädler B. Coenen V.A. Rapid effects of deep brain stimulation for treatment-resistant major depression.Biol Psychiatry. 2013; 73: 1204-1212Abstract Full Text Full Text PDF PubMed Scopus (357) Google Scholar] and OCD [[9]Coenen V.A. Schlaepfer T.E. Goll P. Reinacher P.C. Voderholzer U. Tebartz van Elst L. Urbach H. Freyer T. The medial forebrain bundle as a target for deep brain stimulation for obsessive-compulsive disorder.CNS Spectr. 2016; 493: 1-8Google Scholar,[19]Liebrand L.C. Caan M.W.A. Schuurman P.R. van den Munckhof P. Figee M. Denys D. van Wingen G.A. Individual white matter bundle trajectories are associated with deep brain stimulation response in obsessive-compulsive disorder.Brain Stimul. 2019; 12: 353-360Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar]. It was anatomically characterized [[3]Anthofer J.M. Steib K. Fellner C. Lange M. Brawanski A. Schlaier J. DTI-based deterministic fibre tracking of the medial forebrain bundle.Acta Neurochir. 2015; 157: 469-477Crossref PubMed Scopus (21) Google Scholar,[7]Coenen V.A. Honey C.R. Hurwitz T. Rahman A.A. Mcmaster J. Bürgel U. Mädler B. Medial forebrain bundle stimulation as a pathophysiological mechanism for hypomania in subthalamic nucleus deep brain stimulation for Parkinson's disease.Neurosurgery. 2009; 64 (discussion 1114–5): 1106-1114Crossref PubMed Scopus (76) Google Scholar,[8]Coenen V.A. Panksepp J. Hurwitz T.A. Urbach H. Mädler B. Human medial forebrain bundle (MFB) and anterior thalamic radiation (ATR): imaging of two major subcortical pathways and the dynamic balance of opposite affects in understanding depression.J Neuropsychiatry Clin Neurosci. 2012; 24: 223-236Crossref PubMed Scopus (209) Google Scholar,[10]Coenen V.A. Schlaepfer T.E. Maedler B. Panksepp J. Cross-species affective functions of the medial forebrain bundle-implications for the treatment of affective pain and depression in humans.Neurosci Biobehav Rev. 2011; 35: 1971-1981Crossref PubMed Scopus (181) Google Scholar,[11]Coenen V.A. Schumacher L.V. Kaller C. Schlaepfer T.E. Reinacher P.C. Egger K. Urbach H. Reisert M. The anatomy of the human medial forebrain bundle_ Ventral tegmental area connections to reward-associated subcortical and frontal lobe regions.NeuroImage: Neuroimage Clin. 2018; 18: 770-783Crossref PubMed Scopus (51) Google Scholar,[22]Schlaepfer T.E. Bewernick B.H. Kayser S. Hurlemann R. Coenen V.A. Deep brain stimulation of the human reward system for major depression--rationale, outcomes and outlook.Neuropsychopharmacology. 2014; 39: 1303-1314Crossref PubMed Scopus (100) Google Scholar,[27]Zacharopoulos G. Lancaster T.M. Bracht T. Ihssen N. Maio G.R. Linden D.E.J. A hedonism hub in the human brain.Cerebr Cortex. 2016; 26: 3921-3927Crossref PubMed Scopus (11) Google Scholar] (Fig. 1, right) together with a scientific framework, including its relation to other target regions for OCD and major depression (ALIC, NAC= Nucleus Accumbens, amSTN) [9Coenen V.A. Schlaepfer T.E. Goll P. Reinacher P.C. Voderholzer U. Tebartz van Elst L. Urbach H. Freyer T. The medial forebrain bundle as a target for deep brain stimulation for obsessive-compulsive disorder.CNS Spectr. 2016; 493: 1-8Google Scholar, 10Coenen V.A. Schlaepfer T.E. Maedler B. Panksepp J. Cross-species affective functions of the medial forebrain bundle-implications for the treatment of affective pain and depression in humans.Neurosci Biobehav Rev. 2011; 35: 1971-1981Crossref PubMed Scopus (181) Google Scholar, 11Coenen V.A. Schumacher L.V. Kaller C. Schlaepfer T.E. Reinacher P.C. Egger K. Urbach H. Reisert M. The anatomy of the human medial forebrain bundle_ Ventral tegmental area connections to reward-associated subcortical and frontal lobe regions.NeuroImage: Neuroimage Clin. 2018; 18: 770-783Crossref PubMed Scopus (51) Google Scholar,[24]Schoene-Bake J.-C. Parpaley Y. Weber B. Panksepp J. Hurwitz T.A. Coenen V.A. Tractographic analysis of historical lesion surgery for depression.Neuropsychopharmacology. 2010; 35: 2553-2563Crossref PubMed Scopus (67) Google Scholar] (Fig. 2). More than 50 patients have already undergone DBS to this fiber pathway using individual connectomic imaging for targeting across both indications and 30 patients are already published [[6]Coenen V.A. Bewernick B.H. Kayser S. et al.Superolateral medial forebrain bundle deep brain stimulation in major depression: a gateway trial.Neuropsychopharmacology. 2019; 26: 587Google Scholar,[13]Fenoy A.J. Schulz P.E. Selvaraj S. Burrows C.L. Zunta-Soares G. Durkin K. Zanotti-Fregonara P. Quevedo J. Soares J.C. A longitudinal study on deep brain stimulation of the medial forebrain bundle for treatment-resistant depression.Transl Psychiatry. 2018; 8: 1-11Crossref PubMed Scopus (44) Google Scholar,[23]Schlaepfer T.E. Bewernick B.H. Kayser S. Mädler B. Coenen V.A. Rapid effects of deep brain stimulation for treatment-resistant major depression.Biol Psychiatry. 2013; 73: 1204-1212Abstract Full Text Full Text PDF PubMed Scopus (357) Google Scholar].Fig. 2A, Surgical targets related to the “tract target” taken from Li et al. [[18]Li N. Baldermann J.C. Kibleur A. et al.Toward a unified connectomic target for deep brain stimulation in obsessive-compulsive disorder.2019: 1-28Google Scholar]. B, the slMFB taken from Ref. [[11]Coenen V.A. Schumacher L.V. Kaller C. Schlaepfer T.E. Reinacher P.C. Egger K. Urbach H. Reisert M. The anatomy of the human medial forebrain bundle_ Ventral tegmental area connections to reward-associated subcortical and frontal lobe regions.NeuroImage: Neuroimage Clin. 2018; 18: 770-783Crossref PubMed Scopus (51) Google Scholar]. C, surgical targets for OCD (yellow) related to the slMFB (dark green) taken from Ref. [[9]Coenen V.A. Schlaepfer T.E. Goll P. Reinacher P.C. Voderholzer U. Tebartz van Elst L. Urbach H. Freyer T. The medial forebrain bundle as a target for deep brain stimulation for obsessive-compulsive disorder.CNS Spectr. 2016; 493: 1-8Google Scholar]. The anterior thalamic radiation (ATR, orange) which runs parallel to slMFB in ALIC might in part explain the diverging results from the previous publication [[5]Baldermann J.C. Melzer C. Zapf A. et al.Connectivity profile predictive of effective deep brain stimulation in obsessive-compulsive disorder.Biol Psychiatry. 2019; 85: 735-743Abstract Full Text Full Text PDF PubMed Scopus (95) Google Scholar].View Large Image Figure ViewerDownload Hi-res image Download (PPT) For OCD we fear that a potential next step will be: The pathway in question will live in a digital open source library [[16]Horn A. Li N. Dembek T.A. et al.Lead-DBS v2: towards a comprehensive pipeline for deep brain stimulation imaging.Neuroimage. 2019; 184: 293-316Crossref PubMed Scopus (249) Google Scholar] as a “tract target” atlas (in MNI space) and might be used for individual targeting. Recent literature [[4]Aviles-Olmos I. Kefalopoulou Z. Tripoliti E. et al.Long-term outcome of subthalamic nucleus deep brain stimulation for Parkinson's disease using an MRI-guided and MRI-verified approach.J Neurol Neurosurg Psychiatry. 2014; 85: 1419-1425Crossref PubMed Scopus (112) Google Scholar,[6]Coenen V.A. Bewernick B.H. Kayser S. et al.Superolateral medial forebrain bundle deep brain stimulation in major depression: a gateway trial.Neuropsychopharmacology. 2019; 26: 587Google Scholar,12Fenoy A.J. Schiess M.C. Deep brain stimulation of the dentato-rubro-thalamic tract: outcomes of direct targeting for tremor.Neuromodulation. 2017; 20: 429-436Crossref PubMed Scopus (54) Google Scholar, 13Fenoy A.J. Schulz P.E. Selvaraj S. Burrows C.L. Zunta-Soares G. Durkin K. Zanotti-Fregonara P. Quevedo J. Soares J.C. A longitudinal study on deep brain stimulation of the medial forebrain bundle for treatment-resistant depression.Transl Psychiatry. 2018; 8: 1-11Crossref PubMed Scopus (44) Google Scholar, 14Fenoy A.J. Schulz P. Selvaraj S. Burrows C. Spiker D. Cao B. Zunta-Soares G. Gajwani P. Quevedo J. Soares J. Deep brain stimulation of the medial forebrain bundle: distinctive responses in resistant depression.J Affect Disord. 2016; 203: 143-151Crossref PubMed Scopus (64) Google Scholar,[23]Schlaepfer T.E. Bewernick B.H. Kayser S. Mädler B. Coenen V.A. Rapid effects of deep brain stimulation for treatment-resistant major depression.Biol Psychiatry. 2013; 73: 1204-1212Abstract Full Text Full Text PDF PubMed Scopus (357) Google Scholar] is suggestive of a potential (and plausible) superiority of exclusive direct imaging as an individualized targeting rationale. With respect to this literature it is important to remember that the approaches mentioned by Li et al. [[18]Li N. Baldermann J.C. Kibleur A. et al.Toward a unified connectomic target for deep brain stimulation in obsessive-compulsive disorder.2019: 1-28Google Scholar] have never been contrasted in a direct prospective comparison format. While retrospective analysis (including the use of normative data) can inform future research directions, surgical decision making in centers around the world still remains a non-standardized and expertise-driven affair. Until robust data to the contrary becomes available, anatomical information (structural and connectivity data) used for stereotactic surgery should be individual, identified in a scientific and anatomical framework, personalized and of determined accuracy, with full understanding of the pitfalls and limitations of the technique [[2]Akram H. Hariz M. Zrinzo L. Connectivity derived thalamic segmentation_ Separating myth from reality.Neuroimage: Neuroimage Clin. 2019; : 101758Crossref PubMed Scopus (9) Google L. in precision stereotactic Neurol 2012; Scholar]. scientific information can be available in un-reviewed preprint social and digital should be cautious with such publication these may in use of their un-reviewed This is important in the field of and even more in surgical

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.