The goal of this clinical trial is to evaluate whether a dimensional adaptation of Good Psychiatric Management (called GPM-extended) is more effective than classic Good Psychiatric Management (GPM) in treating adult patients with borderline personality disorder (BPD). Briefly, the GPM-extended model integrates elements from existing adaptations of GPM for narcissistic and obsessive-compulsive personality disorders. It aims to provide a more personalized and dimensional approach to treatment, tailored to each patient's specific personality dysfunction and interpersonal triggers.

The main questions it aims to answer are:

• Does GPM-extended improve overall BPD symptom severity more than classic GPM after one year of treatment?
• Does GPM-extended lead to better outcomes in related clinical domains (e.g., personality functioning, emotional regulation, social functioning...) ?

Researchers will compare two groups:

• Patients treated in a center using the GPM-extended program.
• Patients treated in a center using the classic GPM program. In each group, patients will receive weekly outpatient psychiatric care for one year.

In terms of evaluation, patients will be evaluated at baseline, 4 months, 8 months, and 1 year. They will undergo both clinician-administered and self-report assessments to measure BPD symptoms and other relevant psychological dimensions.

This study hopes to contribute to the development of dimensional evidence-based treatments for personality disorders.

Personality disorders (PD) are among the most common psychiatric disorders, the most studied being borderline personality disorder (BPD), a disorder that affects approximately 1.6% of the general population and is characterised by significant difficulties in emotion regulation, identity and interpersonal relationships. Currently, categories of PD are criticised and many authors have highlighted the need for a more dimensional assessment of PD, using the association of a general factor (described as the level of personality functioning, assessed on personal and interpersonal functioning) with several personality traits used to describe specific personality characteristics (negative affectivity, antagonism/dissociality, detachment, disinhibition, anankastia, psychoticism). The two main models are the DSM-5 alternative personality disorder model (AMPD) and the International Classification of Diseases 11th edition personality disorder model (ICD-11). In particular, several works have suggested that the BPD criteria are one of the most important markers of general personality functioning. While these new models offer new and very exciting possibilities in terms of diagnostic assessment, they have struggled to spread across clinical services ; and, to date, no evidence-based treatments have been developed from these models, limiting their usefulness. Furthermore, these models are also limited by the nature of personality traits (Criteria B), as these represent rigid and stable patterns of dysfunction that may be difficult to represent the complex day-to-day fluctuations in internal psychic coherence and interpersonal functioning characteristic of PD.

One of the most recent treatments for BPD is Good Psychiatric Management. This has proved as effective as specialist therapies such as dialectical behaviour therapy and has also been developed for other personality disorders (notably narcissistic and obsessional-compulsive PD). Each adaptation is based on a specific conceptualisation designed to represent the main ways in which an individual may dysfunction in the personal and interpersonal domains. According to these conceptualisations, each of these three personality disorders presents a specific trigger for its difficulties: threat to relational dependence with fear of rejection and abandonment for borderline PD, threat to self-esteem for narcissistic PD, threat to ability to control for obsessive PD. Thus, some authors have suggested that the development of an adaptation of the GPM incorporating both the central aspects of the dimensional models, but also each of these different triggers in a non-exclusive manner (as they may be found to a greater or lesser extent in each patient suffering from PD), could be both feasible and useful, in particular to resolve the above-mentioned problems.

Indeed, like traditional dimensional models, GPM offers the possibility of a dimensional approach, with personality functioning assessed by the presence or absence of the BPD criteria, and features of personal and interpersonal dysfunction considered holistically using GPM's trigger-based approach. However, unlike traditional dimensional models, GPM has been empirically tested and found to be effective in treating patients who meet the criteria for BPD. In addition, it offers an approach to personality characteristics that is simpler, easier to understand, more accessible to psychoeducation and closer to patients' everyday experience than the personality traits classically used in dimensional models. In addition, although each adaptation of the GPM focuses on different PDs, much of the content remains the same: making and announcing the diagnosis, psychoeducation, case management, recurrent assessment of progress and reassessment if there is no response, multimodal approach including psychodynamic and behavioural psychotherapy, anticipation of crises, and management of symptomatic medication, etc. This may be linked to the fact that, although each disorder has specific triggers/traits, the underlying level of personality functioning is represented by BPD criteria and is therefore expected to be treated by the same psychotherapeutic content. Thus, a dimensional adaptation of GPM seems both relevant and feasible to address the problems of conventional dimensional models, namely the lack of existence of evidence-based treatments associated with these models, and the aspecific nature of personality trait-based approaches.

Altogether, we developed a dimensional adaptation of the GPM (GPM-extended), aiming to treat patients fulfilling the criteria for BPD dimensionally by incorporating elements from the adaptations for narcissistic and obsessional personality disorders. In terms of content, GPM-extended takes the common part of the treatment from the three adaptations and uses it as a basis, while also offering the possibility of constructing treatment goals and exposure targets that are much more specific to a given patient, in particular by carrying out an initial assessment and prioritisation of the various triggers. If this adaptation were to be shown to be effective, it would ultimately improve the diagnostic assessment and management of patients fulfilling BPD criteria, by offering treatment that is more tailored to each profile.

Inclusion Criteria:

• Adults (≥18 years old)
• Diagnosis of borderline personality disorder according to the SCID-II (≥5 out of 9 criteria)
• Provision of written informed consent
• Affiliated with or beneficiary of the French national health insurance system

Exclusion criteria

• Age under 18 years
• Presence of a comorbid psychotic disorder, intellectual disability, severe antisocial traits, major substance use disorder incompatible with intensive therapy without abstinence, anorexia nervosa with somatic risk, or bipolar disorder in acute manic phase
• Individuals under legal protection (e.g., guardianship or legal safeguard)
• Individuals unable to cooperate or complete self- or clinician-administered questionnaires
• Individuals not affiliated with or not beneficiaries of the French national health insurance system

De-identified individual participant data (IPD) used in published analyses will be shared through the Open Science Framework (OSF) platform for each resulting publication. This includes the datasets used in the statistical analyses and the corresponding statistical code. A separate protocol paper will be published prior to the main outcome publications. Data and code will be made available to promote transparency and open science, following the publication of each study output.

The protocol paper will be published before the end of 2026. Other IPD and related materials will be made publicly available starting from the first publication of study results, for an indefinite period, to support transparency and open science.

Access to data will be provided via a dedicated OSF link published alongside each article.