In 2020, cancer claimed the lives of approximately 10 million individuals globally, solidifying its position as the leading cause of death [1]. The prevalence of breast cancer, the most commonly occurring malignancy, is on the rise [2]. With the 5-year survival rate for breast cancer reaching as high as 90% [3], extensive research is currently underway to investigate the psychological aspects of the disease. A study revealed that among patients who underwent surgery for breast cancer, the prevalence of psychological problems was between 30% and 47% [4].

There are different types of mood disorders depending on the extent and intensity of mood elevation [5], such as unipolar depressive disorder (UDD) and other specified bipolar and related disorders, bipolar I disorder, and bipolar II disorder. While only depressive episodes occur among individuals with UDD, those with bipolar disorder (BD) experience a hypomanic or manic episode characterized by mood elevation. BD impacts over 1% of the global population, regardless of the nationality, ethnic background, or socioeconomic status [6]. Since it is mainly diagnosed in early adulthood, it directly affects the economically productive population, resulting in significant financial burdens for the society [6]. Besides, it has the lifelong nature of recurrence and chronicity. Therefore, timely detection of the condition is of importance, and yet, misdiagnosis of BD as UDD often occurs because of the difficulty in differentiating between the two disorders [5], which entails a diagnostic delay. Given that many psycho-oncology patients experience mood disorders [7], it is crucial to understand the current status of BD diagnosis among this population.

However, there is a paucity of studies addressing the prevalence of BD in patients with breast cancer diagnosed with a mood disorder and their characteristics. Thus far, only two studies presenting the prevalence of BD and UDD in patients with breast cancer have been published. In a study utilizing the database of the Taiwan National Health Insurance program, 69 and 1,168 patients, for the first time, received a diagnosis of BD and UDD, respectively, after being newly diagnosed with breast cancer between 2000 and 2005 [8]. The prevalence of BD among mood disorder in this study was 5.58%. In another registry study conducted among patients diagnosed with breast cancer in South East England between 2000 and 2009, 131 and 955 patients were diagnosed with BD and UDD, respectively, based on the admission records of the English National Health Service hospitals during the period spanning 3 years before the cancer diagnosis and 1 year after it [9]. In this study, the prevalence of BD in mood disorders was 12.06%. However, to date, no study has used interview-based assessments to make a diagnosis of mood disorders by strictly applying the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) [10] criteria and presented distinguishing characteristics in comparison with UDD.

The purpose of the current study was to examine the prevalence of BD among patients with mood disorders comorbid with breast cancer who visited a psycho-oncology outpatient clinic and to compare the sociodemographic, clinical, and psychological factors between those with BD and UDD.

Among psycho-oncology outpatients with breast cancer, 56 (67.5%) were diagnosed with UDD, and 27 (32.5%) with BD. Most patients with UDD met the criteria for major depressive disorder (MDD) (89.3%) and most with BD had bipolar II disorder (81.5%) (Table 1). For the sociodemographic factors, the mean age of the patients with UDD and BD was 51.59±9.94 years and 44.89±10.17 years, respectively; the patients with BD were more likely to be younger (t=2.856, p=0.005) than those with UDD (Table 2). Besides, the patients with BD were more likely to be employed than those with UDD (odds ratio [OR]=2.793 [the reciprocal of 0.358], χ²=4.623, p=0.032). In the clinical factors, a significant difference was observed only in the history of suicide attempt, with a higher percentage of patients with BD (51.9%) reporting a previous suicide attempt than that of those with UDD (23.6%) (OR=3.479, χ²=6.528, p=0.011). Apart from the mentioned differences in diagnostic categories, no statistical differences were observed between the two groups. Psychological factors did not differ significantly between the two groups (Table 3).

In this study, approximately one-third of psycho-oncology outpatients with breast cancer and a comorbid mood disorder were diagnosed with BD, with bipolar II disorder being the most common subtype. Individuals with BD were more likely to be younger, employed, and have a history of suicide attempts than those with UDD. These findings highlight the considerable prevalence of BD among individuals with both breast cancer and mood disorder, underscoring the importance of recognizing this condition within this population.

Previous studies reported lower BD prevalence in breast cancer patients with mood disorders—5.58% and 12.06% based on national databases and medical records—than in our study (32.5%) [8,9]. However, our finding aligns with reports from psychiatric outpatient settings, where bipolar II disorder accounts for 30%-61% of outpatients experiencing a major depressive episode [17-19], depending on the setting. In a Hong Kong study [20], 35.9% of the patients, none of whom had bipolar I disorder, received a diagnosis of bipolar II disorder with the duration criterion of a hypomanic episode being lowered to 2 days or longer. This subgroup closely resembled our BD diagnostic group, where bipolar I disorder occurred only in one patient, and the prevalence was similar to that of our sample (32.5%). In contrast, when the DSM-IV criteria for a hypomanic episode (defined as lasting ≥4 days) were applied, the prevalence of bipolar II disorder decreased to 20.5%; similarly, in our study, when subthreshold BD—other specified bipolar and related disorders—was excluded, only 27.7% of our sample met the diagnosis criteria for BD (all, but one, with bipolar II disorder). These findings suggest that BD in psycho-oncology outpatients with breast cancer is significantly underdiagnosed, and its actual prevalence may be similar to that in general psychiatric outpatients. On the other hand, the high prevalence of BD may reflect referral bias, as bipolar depressed patients with more severe or treatment-resistant depressive symptoms, or simply those with a known history of BD, are more likely to be referred to the psycho-oncology clinic. The unexpectedly high prevalence of BD underscores the importance of efforts to identify BD in this population. The assumption that patients experience depression solely because of their cancer diagnosis and treatment, leading to a presumed reactive UDD, must be avoided. Instead, thorough evaluation of the diagnostic criteria for hypomanic or manic episodes is crucial when evaluating patients who present with mood symptoms, even during active cancer treatment.

The difficulty of correctly diagnosing BD is well known. Within the first year of seeking treatment, only 20% of patients with BD currently in a depressive episode receive a timely diagnosis for their condition [21]. A study reported a delay of over 10 years before patients with bipolar II disorder or BD, not otherwise specified, received an accurate diagnosis [22]. Misdiagnosis of BD as MDD is common. In a study involving interviews and chart reviews, 40% of the patients identified with BD had once been diagnosed with MDD [23]. Additionally, in a survey conducted among patients with BD, unipolar depression (e.g., MDD) was the most prevalent incorrect diagnosis, accounting for 60% of the cases [21]. Several explanations for such common misdiagnosis exist. First, the criteria for a major depressive episode is identical for both BD and UDD [5]. Second, patients diagnosed with bipolar II disorder show a greater tendency to seek treatment for depressive symptoms rather than for hypomanic or manic symptoms [5]. Third, in a predominantly depressive course, patients with bipolar II disorder spend most of their time experiencing pure depression [24].

Correct diagnosis of BD is critical to prevent serious consequences of inadequate treatment, such as manic switch, accelerated cycling, and suicide. Antidepressants, originally developed to treat MDD, have the potential to induce a manic episode when used for patients with BD [25], which can pose considerable risks to both the patient and those around them. According to a meta-analysis, the overall incidence rate of manic switch is 12.5% [25]. Agents blocking reuptake of noradrenaline, such as tricyclic antidepressants [26] and venlafaxine [27,28], are known to pose a higher risk of treatment-emergent mania than that of selective serotonin reuptake inhibitors. This is particularly problematic considering they are frequently used for the management of cancer-related neuropathic pain [29]. Detecting BD can prevent inappropriate use of certain antidepressant classes. Furthermore, antidepressants have been associated with accelerated cycling. Among the patients who had been misdiagnosed with UDD and prescribed antidepressants, 23% exhibited a new or aggravated rapid-cycling pattern that could be attributed to antidepressant usage [23]. A study on individuals with rapid-cycling mood disorder demonstrated that more than 70% of them were receiving antidepressants at the onset of rapid-cycling episodes, and the persistence of rapid cycling was associated with the use of antidepressants in approximately half of them [30]. Rapid cycling is common among female patients with bipolar II disorder [31], and most BD participants in our sample fit this description. Epidemiological studies report that over 20% of individuals with BD attempt suicide [32]. Among the participants in the Epidemiologic Catchment Area database, the lifetime rates of suicide attempt were almost twice greater in individuals with BD (29.2%) than in those with UDD (15.9%) [33]. A higher risk of suicidal behavior occurs in patients with bipolar II disorder, the most prevalent BD subtype in the current study, than in those with bipolar I disorder or UDD [34]. The misdiagnosis of BD as UDD could lead to a lack of the use of mood stabilizers, and the absence of adequate therapy involving mood stabilizers may result in suboptimal alleviation of symptoms and increased risk of suicide [35-37].

In our study, the mean age of patients with BD was lower than that of those with UDD. Although onset age data were unavailable for our participants, we cautiously speculate that the younger age in BD may reflect the earlier onset of BD (approximately 25 years) compared with UDD [38,39]. Patients with BD were more likely to be employed than those with UDD, possibly reflecting age differences. Moreover, the fact that most patients with BD were employed has an important implication from an economic perspective. Regardless of comorbid breast cancer, BD itself imposes a significant economic burden, including indirect costs (productivity losses and unemployment) and direct costs (psychiatric services, hospitalization, psychotropic medications) [40]. A previous research revealed that patients with bipolar II disorder were incapable of performing their work duties in 20% of the assessed months over 15 years [41]. Therefore, accurate diagnosis and optimal treatment of BD are of great importance.

Patients with BD in the current study were over three times more likely to have a history of suicide attempt compared with those with UDD. This finding aligns with previous research showing that bipolar II disorder carries a higher risk of suicidal behavior than UDD [34]. A history of suicide attempt is a well-established risk factor of suicide [42], and approximately 25%-50% of the patients diagnosed with BD make at least one suicide attempt in their lifetime [43]. Moreover, the maximum lethality of suicide attempts is more likely to be higher for those with BD than for those with MDD [44], with patients with BD reporting greater levels of aggression and impulsivity [44], which are endophenotypes associated with suicide [45]. Therefore, accurate diagnoses of BD are crucial to ensure optimal treatment, as individuals with BD have a higher incidence of suicide attempts than those with UDD. In an epidemiological study, the prevalence of lifetime suicide attempts was reported to be higher in individuals with BD than among those with MDD [33]. Similarly, an 18-month prospective study demonstrated a greater cumulative occurrence of suicide attempts in patients with BD than among those with MDD [46]. Besides, for patients diagnosed with BD, suicide attempts are most commonly observed during a major depressive episode [47,48] or during mixed conditions [47,48]. When BD is misdiagnosed with UDD and suboptimally treated with antidepressants without mood stabilizers or antipsychotics, some patients with BD can develop mixed conditions [49] characterized by agitation, excitation, or mental arousal-anxiety, which has been associated with an increased risk of suicidal behavior [49]. Compared to patients with BD and MDD in a fully remitted state, those with BD in a mixed state have a 65 times higher incidence rate for suicide attempts [46].

No significant differences in the intensity of psychological factors—depression, anxiety, insomnia, suicidal ideation, and functional well-being—were observed between the two diagnostic groups, indicating that thorough psychiatric attention should be provided irrespective of the specific mood disorder diagnosis.

This study has several limitations. First, the sample size was small. Second, the absence of a structured interview for psychiatric diagnosis due to limited human resources, along with the omission of case discussions raises, concerns about the accuracy of the diagnoses. However, all psychiatric diagnoses were made by board-certified psychiatrists specialized in mood disorders or consultation-liaison psychiatry, strictly adhering to the DSM-5 criteria of mood disorder. Moreover, although physical illness-related symptoms in medically ill populations may resemble depressive symptoms and potentially inflate the prevalence of current major depressive episodes, it is unlikely that the absence of a structured interview substantially affected the identification of past hypomanic or manic episodes. Third, as the level of psychological challenges may differ based on the stage of the disease and treatment status among individuals with breast cancer, it is not possible to generalize the current findings as a universal feature of patients with breast cancer who have either UDD or BD. Instead, these findings should be considered as characteristics specific to patients with these illnesses who sought treatment at a psycho-oncology clinic within a tertiary cancer center. Fourth, all participants in the study were ethnic Koreans, and therefore, caution should be exercised when generalizing the findings to other ethnic groups. Fifth, detailed oncological information was not collected, including cancer stage, metastatic status, recurrence history, hormonal receptor subtype, and treatment phase, all of which may influence psychological burden, treatment response, and functional functioning. Sixth, detailed data on psychotropic medication use were not collected, which is an important limitation because such medications may influence depressive severity, sleep patterns, anxiety symptoms, and suicidality; in addition, some cases classified as BD may have included BD diagnosed following antidepressant-induced hypomania or mania, and this possibility cannot be ruled out. Despite these limitations, this study had notable strengths. First, this was the first study to explore the differences in prevalence and characteristics of BD and UDD in a psycho-oncology breast cancer population. Second, comprehensive measurements were utilized to address the multifaceted aspects of psychological issues related to mood disorders.

In conclusion, this study revealed that, among psycho-oncology outpatients with breast cancer and a concurrent mood disorder, approximately one-third were diagnosed with BD, with type II being the predominant subtype. Patients with BD had a higher likelihood of being younger, employed, and having a history of suicide attempts than that of those with UDD. These novel findings underscore the significant occurrence of BD and the importance of making efforts to recognize and address the condition within this population. Considering the exploratory nature of the sample, replication in larger, structured, multi-center studies is warranted to validate and expand upon these findings.