Risk of major depressive disorder in adolescent and young adult cancer patients in Japan

Abstract Objective To estimate the risk of major depressive disorder (MDD) in adolescent and young adult (AYA) patients with cancer in Japan and identify risk factors for MDD among these patients. Methods This was a matched cohort study using a large claims database in Japan. Included patients were aged 15–39 years, newly diagnosed with cancer during 2012–2017 and assessable for a follow‐up period of 12 months. Kaplan–Meier estimates and Cox proportional hazards models were used to calculate hazard ratios (HR) and 95% confidence intervals (CI) for MDD in the AYA patients with cancer versus age‐, sex‐ and working status‐matched cancer‐free controls. A subgroups analysis of the AYA patients with cancer was performed to explore MDD risk factors. Results A total of 3559 AYA patients with cancer and 35,590 matched controls were included in the analysis. Adolescent and young adult patients with cancer had a three‐fold higher risk for MDD compared with cancer‐free controls (HR, 3.12; 95% CI, 2.64–3.70). Among cancer categories with >100 patients, patients with multiple cancer categories, including those with metastatic cancer (HR, 6.73, 95% CI, 3.65–12.40) and leukemia (HR, 6.30; 95% CI, 3.75–10.58), had the greatest MDD risk versus matched controls. Patients who received inpatient chemotherapy as initial treatment had a higher risk for MDD than patients without chemotherapy (HR, 0.43; 95% CI, 0.30–0.62). Conclusions Adolescent and young adult patients in Japan with cancer are at high risk for MDD. Particularly, those with multiple cancer categories, leukemia, and those who receive aggressive anticancer treatments should be monitored closely for symptoms of MDD.


| BACKGROUND
Cancer is a serious and potentially life-threatening illness that can have profound effects on mental health. 1,2 Most patients diagnosed with cancer experience a normal albeit painful emotional reaction; however, some develop clinical depression. Cancer patients with depression have poor cancer treatment adherence, increased healthcare utilization (including hospitalizations), and elevated risks for mortality and suicide. [3][4][5][6] Literature reviews have reported that 5%-10% of adult patients with cancer (aged ≥18 years) have experienced a major depressive episode (a rate 2-3 times higher than in the general population). 7,8 Additionally, observational studies and reviews have reported prevalence rates of depressive symptoms of up to 55% among adult patients with cancer (up to four times higher than the general population). 1,[8][9][10][11][12][13] A nationwide matched-cohort study conducted in Sweden found that the risk of mental disorders peaked during the first week after cancer diagnosis and that depression had the highest cumulative incidence among the various studied mental disorders (including stress reaction or adjustment disorder, substance abuse, or anxiety, somatoform or conversion disorders). 14 16) compared with cancer-free matched controls during an observation period from 6 months before to 12 months after cancer diagnosis. 15 MDD risk was highest in patients with multiple cancer categories, most of whom had metastatic cancer, pancreatic cancer, and brain cancer. In the study mentioned above, 10.6% of included patients were aged ≤39 years, who are considered adolescent and young adults (AYAs). 16 Adolescent and young adults (individuals aged 15-39 years) have limited access to dedicated institutions for healthcare and research. 17 Hence, the risk of cancer and adverse cancer outcomes may be underestimated in this age group, which may have contributed to a lack of improvement in cancer survival rates over the years. Adolescent and young adult patients with cancers are at higher risk for depression compared with older adults. 12 Observational studies from Asia, Germany, and the USA have reported that 25%-39% of AYA patients with cancer experience depressive symptoms. [18][19][20] Data from Canada showed that AYA survivors of cancer had increased outpatient mental health visits and a higher risk of a severe psychiatric episode compared with matched controls. 21 Furthermore, AYA patients with cancer carry a disproportionate burden of distress and psychological symptoms compared with older patients with cancer. 12 A survey of 575 AYA patients with cancer revealed that although 90% of patients reported at least one emotional concern (66% reported depression or loss of interest in daily activities), only 43% of those with concerns sought help. 22 Barriers to help-seeking included not wanting to ask, embarrassment, and being told that their emotions were "normal." Among those who sought help, 37% had difficulties obtaining assistance.
Despite previous findings, more robust data on the real burden of MDD among AYA patient with cancers, especially in comparison with cancer-free controls and using large sample sizes, remains limited. 23 A 2020 systematic review and meta-analysis of psychiatric disorders in AYA cancer survivors could include only four and three studies, respectively, from 7934 studies screened. 16 There are also significant knowledge gaps regarding the prevalence of MDD in AYA patients with different cancer categories and the various demographic and clinical variables associated with MDD risk in these patients. Such information would be essential in providing appropriate support for AYA patients with cancer. This study aimed to estimate MDD risk in AYA patients with cancer aged 15-39 years in Japan compared with matched cancer-free controls.

| Study design and setting
This was a retrospective cohort study that included AYA patients with cancer in Japan, each matched to 10 cancer-free controls.
The ratio of 1:10 was in line with that used in a nationwide matched cohort study in Sweden. 14 Data were retrieved from an administrative claims database housing multiple health insurance societies in Japan (JMDC database; JMDC Inc.). 24,25 The JMDC database is an anonymized database on insured medical services and prescriptions. In Japan, all individuals are covered by either a national or an employer-based comprehensive health insurance scheme. 26 The JMDC database covers employees of medium to large companies and their non-working family members aged  Classification of Diseases and Related Health Problems, 10th revision (cancer: C00-C95), as listed in Table S1. 28 Patients with diagnoses of multiple cancer categories were defined as those having ≥2 ICD-10 cancer categories.

| Study population
The study had an 18-month observation period starting from 6 months before through 12 months after the month of cancer diagnosis ( Figure S1). Exclusion criteria included any history of depression within 6 months before the start of the 18-month observation period (window period), patient age <15 or ≥40 years, and having discontinuous health insurance enrollment (due to e.g., change of employer) before or after 12 months of the index months (study period).
Cancer-free controls included those with no diagnosis of cancer during the study period and no history of depression during the window period.

| Variables and assessment
For AYA patients with cancer, the following baseline characteristics were recorded: age, sex, working status (working vs. non-working), category of cancer diagnosis, and initial treatment in the first 2 months of cancer diagnosis (chemotherapy [none, inpatient, or outpatient], radiation therapy [none, brachytherapy, or external beam radiotherapy], and presence or absence of surgery with ≥5 days' hospitalization). All AYA patients with cancer and matched controls were followed up for incident MDD, defined as either an ICD10 code of F32 (depressive episode) or F33 (MDD, recurrent) during the observation period.

| Statistical analyses
Categorical variables were reported as counts and proportions whereas continuous variables were reported as means, standard deviations, medians, and ranges. A two-sided significance level of 5%

| RESULTS
A total of 3559 patients aged 15-39 years, with newly diagnosed cancer during the enrollment period and no recent history of depression were included in the AYA cancer group ( Figure S2). The age-matched cancer-free control group included 35,590 individuals.
There were more female than male AYA patients with cancer (59.9% vs. 40.1%) and more than half were full-time workers (59.5%; Table 1). In both the cancer and control groups, 87% were aged ≥25 years. In the AYA patients with cancer, the most common can-  These results may reflect more aggressive tumor behaviors, diagnosis at advanced stages, and the administration of more aggressive treatments for these cancers in AYA patients. 23,33 Given the elevated MDD risk in these patients, special supportive measures should be considered.
Cancers with better prognoses (e.g., thyroid cancer) were generally associated with lower MDD risk, although patients with breast cancer, which has a 5-year relative survival rate and conditional 5-year relative survival rate of ≥90%, 34 still had a four-fold higher risk for MDD compared with matched controls. Several factors may contribute to this finding with breast cancer, including longterm complications such as lymphedema, treatment-related fatigue, and cognitive impairment, adverse effects of long-term hormonal therapy, and sexual dysfunction. 35 Similarly, patients with ovarian cancer experience unique psychological stresses, such as reduced sexual activity and satisfaction; fertility concerns; anxiety arising from the genetic association of the disease which could lead to fear for their female offspring; and the usual trajectory of ovarian cancer, which includes aggressive treatments and recurrence. [36][37][38] Among the AYA patients with cancer, there was a significant 57% lower risk for MDD among those who did not receive chemotherapy as initial therapy compared with those who received inpatient chemotherapy. This finding mirrors the data in adult matched patients not receiving initial chemotherapy (HR, 0.53; 95% CI, 0.46-0.62). 15 The increased risk of MDD among those who had inpatient chemotherapy may be related to the advanced stage of their disease.
Although the prevalence of mental disorders in Japan is lower than in Western countries, 39

| Study limitations
To date, this study represents the most comprehensive assessment of the burden of MDD in AYA patients with cancer in Japan. However, it also has some limitations. In the claims database, diseases claimed without clinical definitions may lead to overestimation of their be interpreted with this in mind. An important additional consideration is the potential for underdiagnosis or a lack of MDD symptom disclosure due to high levels of mental health stigma in Japan, including amongst AYA populations. 43 Lastly, the number and type of confounders were limited by the availability of data from the anonymized administrative claims.

| Clinical implications
There is a need for physicians treating cancer patients to be aware of the high possibility of MDD among AYA patients. As MDD among AYA cancer patients is treatable, it is crucial to look out for depressive symptoms, especially in patients whose cancer has a poor prognosis or are receiving chemotherapy during hospitalization, and ensure that these patients receive appropriate support and treatment. In consideration of the high risk in AYA patients, and their limited access to and use of dedicated mental health resources, a special support system potentially including depression screening, multidisciplinary care, or increased specialization in managing AYA patients may be desirable in Japan.

| CONCLUSIONS
In Japan, AYA patients with cancer had an increased risk for MDD compared with cancer-free controls. At highest risk were those with multiple cancer categories, brain/CNS cancer, leukemia, and stomach cancer. These patients and those who received aggressive treatment should be monitored closely and provided appropriate support and mental health care.

ACKNOWLEDGMENTS
Medical writing support was provided by Oliver Chew and Ivan Olegario of MIMS Pte Ltd., which was funded by Takeda   Izumi Mishiro and Shinji Fujimoto report personal fees from Takeda during the conduct of the study and outside the submitted work.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from JMDC Inc. but were used under license for the current study; therefore, restrictions apply and the data are not publicly available.
For inquiries about access to the data set used in this study, please contact JMDC (https://www.jmdc.co.jp).