Paradigm shift on the concept of schizophrenia that matches with both academic and clinical needs

Since its inception over 100 years ago, the boundary of the diagnostic category “schizophrenia” has been debated (Carpenter, 2013). More fundamentally, it has been questioned whether this group forms an independent disease entity or a syndrome. The current consensus among scholars is that schizophrenia is a syndrome and includes heterogeneous subsets of patients with unknown etiology (Insel, 2010). Studying schizophrenia as a single disease entity impedes further mechanistic study and development of new treatment. To overcome this dilemma, many experts now propose to revisit (or even deconstruct) the concept of schizophrenia and introduce a dimensional approach to dissect the patient group (Owen et al., 2016). In this article, we will discuss how we can consider the paradigm shift to match with both academic and clinical needs.

We are a group consisting of psychiatrists (physician scientists) and scientists of molecular neuroscience. From our standpoint, we also agree that the heterogeneity is a major problem that impedes better understanding of disease mechanism and pathophysiology (Guloksuz and van Os, 2018). In addition, the multifactorial nature of the etiological factors also enhances the difficulty in addressing causal mechanism in the disease pathophysiology (Namkung et al., 2018). We have seen that scholars in the two camps of clinical practice and basic science may sometimes view the same topics from different perspectives. The difference can provide complementary viewpoints and be constructive a multifaceted approach, whereas this may elicit misunderstanding with each other. For example, the outcome of psychiatric genetics is viewed as a conceptual challenge to classic disease categories, such as schizophrenia and bipolar disorder (Owen, 2014), whereas some scholars still regard that genome-wide genetic studies have underscored genes for each disease, such as the genes for schizophrenia. Success in multifaceted research may be crucial for the paradigm shift described above. Thus, the “mind the gap” concern (the possible gap described above) is an issue that cannot be overlooked, particularly in professional education (Diester et al., 2015).

When we consider the potential paradigm shift on the concept of schizophrenia, we regard a chapter written by William Carpenter in 2013 as a good guideline (Carpenter, 2013). In this chapter, he proposed four strategies for the paradigm shift under discussion: (1) identifying patient subgroups to enhance the homogeneity; (2) deconstructing schizophrenia and identifying key domains of psychopathology; (3) deconstructing schizophrenia at the levels of neural circuits and behavioral constructs; and (4) considering stages of both vulnerability development prior to onset and disease progression. We agree with all four strategies as potentially promising routes for the paradigm shift, but we also acknowledge that how we weigh these four may be very important between academic advancement and daily needs of clinical setting. For example, although the classic categorical approach for schizophrenia may limit academic research, this approach still exceeds reliability and clinical utility when compared with new dimensional approaches. Before fully implementing a new dimensional approach to clinical settings, it will be important to confirm that the new approach is able to uniquely validate course and treatment of a given patient. We can envision that such a dimensional approach in the clinical setting will provide clinicians with diagnostic tools based on commercially available biomarkers helping to identify and screen patients at the onset of the illness. Such approach will result in tailoring treatment to a more homogenous group of patients improving response to treatment and changing the course of illness. Altogether, we expect a gradual and steady transition from a categorical to dimensional emphasis.

Many groups, including ours, have found a reduction of the endogenous antioxidant glutathione (GSH) in peripheral tissues from patients with psychotic disorders such as schizophrenia (Yao and Keshavan, 2011). In our study, we observed that total GSH levels in plasma positively correlated with composite neuropsychological performance (Coughlin et al., 2020). More recently, multiple groups have observed a significant reduction of GSH also in the brain [mainly the anterior cingulate cortex (ACC)] by using the 7-Tesla magnet resonance spectroscopy (MRS) (Sydnor and Roalf, 2020; Wang et al., 2019). However, the levels of brain GSH are similar between control and patient groups at the individual levels, and the difference between these two groups is subtle, even if it is statistically significant. One may hypothesize that the heterogeneity of the patient group affects the data and that a subgroup of the patients may be more specifically associated with the GSH change. Both hypothesis-driven and unbiased approaches may be able to address the putative subgroup. Treatment-resistant schizophrenia is one of the central clinical problems (Howes et al., 2017), and a dopamine-independent mechanism (e.g., glutamatergic deficit) is hypothesized for its pathophysiology (Demjaha et al., 2014). Given that GSH can be a reservoir for synaptic glutamate (Sedlak et al., 2019), it is possible to hypothesize the potential link between treatment resistance and GSH deficits. Indeed, two clinical studies have supported a preferential link of the GSH reduction in the ACC in treatment-resistant first episode psychosis patients (Dempster et al., 2020; Yang et al., 2021). This story depicts the first several steps of a longer journey. However, pinning down the mechanism between GSH deficits and dopamine-independent mechanism(s) for clinical manifestations is a biologically tractable question. Such mechanistic studies by using GSH as a proper biomarker may further purify the subgroup beyond the clinical criteria of treatment resistance that is currently used (Howes et al., 2017). The approach of using biological findings for a better classification is advantageous, because this may lead to a novel mechanism-driven therapy for a precise subgroup of patients defined by proper biomarker(s). In summary, the first strategy that Carpenter described (Carpenter, 2013) may be a realistic route for the constructive paradigm shift.

We anticipate that many other chapters within this special issue on the topic of “Re-inventing schizophrenia: updating the construct” will discuss the second and third strategies (Carpenter, 2013), in conjunction with the struggle of revising Diagnostic and Statistical Manual of Mental Disorders, 5th Edition (DSM-5) and introduction and refinement of the Research Domain Criteria (RDoC). Although these efforts of deconstructing schizophrenia by dimensional approaches are essential, which are frequently considered in a cross-sectional context, we should not forget that the pathophysiology of schizophrenia is also dynamically changed in a longitudinal manner. In short, schizophrenia is syndromic and heterogeneous in both cross-sectional and longitudinal viewpoints. Thus, the forth strategy that considers distinct stages from vulnerability development to disease progression (Carpenter, 2013) is particularly important. A famous effort in this line is the staging model proposed by McGorry (McGorry et al., 2006). This longitudinal perspective can contribute to the efforts of prophylactic intervention and early-stage treatment, and may also reshape pathophysiology-based patient stratification after onset of the disease. Recent studies have shown that the occurrence of symptom exacerbation in early-stage psychosis could result in brain changes, which is separate from the impact of antipsychotic medications (Andreasen et al., 2013; Voineskos et al., 2020). These observations are consistent with the study that reports differences of brain connectivity between patients in different stages, associated with the occurrence of symptom exacerbation (Griffa et al., 2019). Addressing the heterogeneity within schizophrenia in association with disease progression is directly connected with brain pathophysiology and mechanism, which in turn is crucial for mechanism-driven treatment strategies to a more homogeneous subgroup.

In summary, we emphasize the significance of overcoming heterogeneity within the syndromic name of schizophrenia, by applying dimensional approaches and scientific mechanism-driven efforts in both cross-sectional and longitudinal manners. Some scholars anticipate that in 30 years the name of schizophrenia may be viewed as a past cornerstone, and will not be used in psychiatry. Although we do not disagree with the paradigm shift in principle, we expect that this shift should happen gradually but steadily by balancing the needs of both clinical setting and academic science. Through the efforts of stratifying (and gradually deconstructing) the concept of schizophrenia in precise scientific approaches, we are optimistic that several therapeutic strategies will be developed for specific features of “current” schizophrenia, in particular those that affect prognosis and social function of the patients.