Introduction

A greater understanding of the molecular and immunologic processes driving cancer contributes to continued innovation of cancer treatment, while rising costs of the treatment has the potential to increase the risk of financial hardship [1]. Financial toxicity (FT) is a term proposed to describe the impact of the cost of cancer care at patient-level [2], referring to a potential consequence of subjective financial distress that patients experience owing to cancer-related treatment costs [3]. FT is a unique term used to consider financial complaints as being of the same nature as physical toxicity [2]. FT was reported in 28–48% of cancer survivors using monetary measures and in 16–73% of cancer survivors using subjective measures [4]. Molecular-targeted drugs and immune checkpoint inhibitors have been identified as risk factors for FT [5]; however, FT due to molecular-targeted and immune therapies has not been thoroughly explored. The impact of FT is diverse and includes personal-related outcomes, such as lifestyle changes, poor quality of life, and bankruptcy, as well as disease-related outcomes, such as the abandonment, delay, and discontinuation of treatment and, in serious cases, an increased risk of mortality [6]. Even in publicly funded health care systems, FT can potentially lead to negative outcomes for patients with cancer [7]. National health insurance system in Japan reduces out-of-pocket expenses for insured medical care to 30% of the total treatment cost or 10–20%, depending on specific conditions including patients’ age and income. In addition, the High-Cost Medical Expense Benefit System further reduces monthly out-of-pocket expenses when utilized. Despite these mitigating systems, FT is a substantial issue in Japan [8, 9]. A systematic review supports the notion that public healthcare insurance does not necessarily protect against FT [10].

Patient-physician discussions about costs were limited to 24% in oncology settings, even when broadening the definition of cost conversation to its most liberal interpretation, and the rate of cost conversations was lower than that in non-oncology settings [11]. Despite the fact that 52% of patients with cancer wanted to talk to their oncologist about treatment-related costs, only 19% actually did [12]. Only half of the oncologists were aware of the subjective burden their patients experienced, which strongly suggests that oncology professionals underestimate the subjective financial burden [13]. For oncology professionals to detect FT in cancer care before FT gets worse, they should pay closer attention to patient experiences to find clues. The impacts of financial hardship on patients with cancer and their families appear on various domains, including family finances, lifestyle, psychological well-being, health, spiritual well-being, and family dynamics [14]. Parallelly, patients facing FT try to address financial burden using mitigating strategies such as making financial and emotional adjustments and seeking support and resources to cope [15]. The relationship between FT and patient experience is also theoretically consistent with the framework that objective financial burden manifests as FT through subjective financial distress [3]. Turning to patient experiences could aid in enhancing our sensitivity to FT and referrals to appropriate financial resources.

Therefore, this study aimed to evaluate the FT of patients with cancer, undergoing outpatient chemotherapy with molecular-targeted drugs and immune checkpoint inhibitors, and to explore the independent factors related to FT for clinical assessment.

Patients and methods

Patients

We recruited cancer outpatients from nine hospitals in Japan. Inclusion criteria included continuing a regimen containing any molecular-targeted agent or immune checkpoint inhibitor as cancer treatment for ≥ 2 months (combination of cytotoxic therapy was not considered), being informed of his/her cancer diagnosis, being 20–60 years of age at the time of the survey request, and being able to fill out the questionnaire independently. The age limit was set because Japanese people usually experience change in income due to retirement (and being rehired in some cases) at the age of 60, and people over the age of 70 who meet the requirements pay less out-of-pocket treatment expenses owing to medical care system for older individuals. Exclusion criteria included receiving welfare benefits at the time of the survey request and receiving each dose of medication under inpatient care.

Procedures for data collection

An anonymous, self-administered questionnaire was administered from October 2021 to August 2022. This study was approved by the Ethical Review Board of Osaka University Hospital (No. 21267 (T7)) and the participating hospitals. Nominated clinical nurses from each hospital selected research candidates in a convenient way and briefed them on the study. Candidates gave verbal consent prior to their inclusion. The included participants were given questionnaire forms along with a return envelope addressed to the institution of the research representative, were instructed to fill out the questionnaire themselves, and provided a checkbox to express their consent at the beginning of the questionnaire. The returned questionnaire with an entry in this consent section was considered as consent to participate.

Measurements

Financial toxicity

FT was evaluated using the Japanese translation of the COmprehensive Score for Financial Toxicity (COST) [8, 9]. The original version of the COST [16] was developed to assess FT in patients with cancer, and its reliability and validity have been confirmed [17]. The COST is an 11-item scale, and respondents rate each item for the previous week on a 5-point scale ranging from 0 (not at all) to 4 (very much). After inverting the scores of the reversed items, a total score between 0 and 44 points is calculated. A lower score indicates greater FT. A previous study proposed a cutoff of 17.5 point to predict an adverse outcome for quality of life (sensitivity, 0.673; specificity, 0.657) [5].

Experience associated with the financial burden of cancer treatment

Based on a literature review and the nationwide Cancer Patient Experience Survey in Japan, 11 items were extracted [3, 9, 15, 18, 19]. Respondents were asked whether they had experienced each item since starting the cancer treatment they were receiving. If respondents had experienced the items, they rated the degree of their distress on a 3-point scale (0 = not distressed; 1 = distressed; 2 = strongly distressed).

Patient characteristics and information related to financial status

Demographic and clinical data regarding patient characteristics were collected using the questionnaire. We also collected information on expenditures (such as out-of-pocket treatment expenses, frequency of hospital visits for cancer treatment, and subscription to private health insurance) and information related to income (such as employment status, household income, and household savings).

Statistical analysis

All variables were summarized using descriptive statistics. Then, using t-tests, we first compared the COST score according to whether the participants had experienced each item. Second, the correlation between the COST score and the number of items that the participants had experienced was investigated. Third, we compared the COST score by participant characteristics. Correlation coefficients were calculated for continuous variables, and p values were calculated using t-tests (two groups) and one-way analysis of variance (ANOVA) (three or more groups) for categorical variables. When differences among groups were significant in ANOVA, Tukey post-hoc tests or Games–Howell post-hoc tests were conducted according to whether or not equal variances were assumed. Finally, a multiple regression analysis was conducted to explore independent factors associated with a lower COST score. The items that achieved p < 0.05 in univariable analysis were entered by the forward selection method. The significance level was defined as p < 0.05. IBM SPSS statistics, version 28.0 (IBM Japan Ltd., Tokyo, Japan), was used for statistical analysis.

Results

The 174 copies of the questionnaire were distributed to research participants. A total of 149 (85.6%) copies were returned, of which 138 (79.3%) were analyzed. The 11 copies that were excluded were six cases of respondents not meeting the inclusion criteria (ineligible age), three cases that had extremely missing values on the COST measure (none of, or only 1 item responded), and two cases that did not give consent.

Summary of participants

Table 1 shows the participant’s characteristics. Participants were diagnosed with cancer at a mean age of 48.3 ± 7.3 years (mean ± standard deviation), and their mean age at the time of the survey was 51.5 ± 7.1 years. The majority of participants were female (69.6%). Breast cancer was the most common type (43.5%). Most participants received molecular-targeted therapy (71.7%) and continued the treatment without determining its duration (typically, metastatic or recurrent cancer treatment) (57.2%).

Table 1 Clinical and demographic characteristics of the participants (N = 138)
Full size table

Details of the financial status of the participants are shown in Table 2. Approximately half (51.4%) of the participants paid 40,000–60,000 JPY per month on average over the prior 2 months. Subscription to private insurance was widespread (84.1%). A total of 86 (62.3%) participants were employed, but 30 (21.7%) had irregular employment. Figure 1 shows changes in income after a cancer diagnosis. The personal income after cancer diagnosis “decreased significantly” in 54 participants (39.1%) and “decreased slightly” in 29 participants (21.0%), indicating that income had decreased in 60.1% of the participants. Furthermore, 73 participants (52.9%) reported a decrease in household income as well.

Table 2 Financial status of the participants (N = 138)
Full size table
Fig. 1
figure 1

Changes in a personal income and b household income after cancer diagnosis

Full size image

Financial toxicity based on the COST measure

Table 3 shows the COST score that the participants reported. Cronbach’s alpha for the COST score was 0.850. The mean and median total COST scores of the participants were 17.0 ± 8.4 points and 18 points, respectively. A total of 68 (49.3%) participants reported COST scores at the cutoff point or lower. The items with a low mean score in ascending order were “I know that I have enough money in savings, retirement, or assets to cover the costs of my treatment” (0.8 ± 1.0), “I am satisfied with my current financial situation” (0.9 ± 0.9), and “I feel in control of my financial situation” (1.3 ± 1.1).

Table 3 Financial toxicity evaluated using the COmprehensive Score for Financial Toxicity (N = 138)
Full size table

Experience associated with the financial burden of cancer treatment

In terms of frequency of experience, “thinking about how to make ends meet” was the most frequently experienced (99 participants, 71.7%). Although 55 participants (39.9%) experienced “hesitation regarding continuing treatment based on finances,” only 25 participants (18.1%) had the experience of “talking to health care providers about financial worries.” The most common “strongly distressed” experience was “relying on relatives and others to help with the costs of health care” (10/17, 58.8%), followed by “cutting through my deposits and savings” (38/80, 47.5%) and “hesitation regarding continuing treatment based on finances” (24/55, 43.6%).

Relationship between the COST score and the experience associated with the financial burden of cancer treatment

Table 4 compares the COST scores according to prior experience associated with the financial burden of cancer treatment. The COST scores were significantly lower when prior experiences were present compared to when there was no prior experience for all items, except for “taking measures for using public subsidies and support systems.” As shown in Fig. 2, the number of items that the participants experienced was significantly correlated to the COST scores (r = − 0.715, p < 0.001); that is, the COST scores significantly worsened as these experiences accumulated.

Table 4 Differences in the COmprehensive Score for Financial Toxicity by prior experience associated with the financial burden of cancer treatment (N = 138)
Full size table
Fig. 2
figure 2

Pearson correlation coefficient, r =  − 0.715, p < 0.001. The vertical axis depicts scores of the COmprehensive Score for Financial Toxicity (COST), and the horizontal axis depicts the accumulated number of items responded to that participants have experienced

Correlation between financial toxicity and the accumulation of experiences associated with the financial burden of cancer treatment (n = 134)

Full size image

Relationship between the COST score and patient characteristics

Table 5 compares the COST scores by participant characteristics. Regarding clinical and demographic characteristics, the COST scores were significantly lower when participants had comorbidities (p = 0.020) and graduated from junior high school and high school (ANOVA, p < 0.001; Games–Howell post-hoc test, compared to university and graduate school, p < 0.001). Financial status of the participants including mean out-of-pocket treatment expenses per month (p = 0.049), employment status (p = 0.009), annual household income (p < 0.001), and household savings (p < 0.001) were also significantly correlated to the COST scores. Tukey post-hoc tests indicated that lower COST scores were significantly related to having paid 60,000 JPY or more (compared to 20,000 to 40,000 JPY, p = 0.045), being on leave or retired due to cancer (compared to employed, p = 0.007), having less than 4 million JPY as annual household income (compared to 4 million to 8 million, p = 0.004; to 8 million or more, p < 0.001), and having less than 4 million in household savings (compared to 4 million to 8 million, p = 0.048; to 8 million or more, p = 0.001).

Table 5 Differences in the COmprehensive Score for Financial Toxicity by participant characteristics (N = 138)
Full size table

Factors affecting the COST scores in multiple regression analysis

Table 6 shows the results of multiple regression analysis with the COST score as the dependent variable. Factors independently related to the COST score were “hesitation regarding continuing treatment based on finances” (sβ = − 0.410, p < 0.001), “cutting through my deposits and savings” (sβ = − 0.253, p = 0.003), and “reducing spending on basics like food or clothing” (sβ = − 0.205, p = 0.046) in prior experience associated with the financial burden of cancer treatment, and having comorbidities (sβ = -0.156, p = 0.032) as clinical characteristics (analysis of variance, p < 0.001; adjusted R2 = 0.545; Durbin–Watson ratio = 1.962). The variance inflation factor ranged from 1.059 to 2.130.

Table 6 Independent factors related to the COmprehensive Score for Financial Toxicity (n = 95)
Full size table

Discussion

To the best of our knowledge, this is the first study to evaluate the relationship between FT and patient experiences associated with financial burden of molecular-targeted and immune therapies. The main findings of this study are as follows: (1) one in two participants experienced a high level of FT that might have reduced their quality of life; (2) coping strategies to raise funds for treatment expenses was significantly related to higher FT; and (3) participants with higher FT hesitated more in continuing treatment based on finances.

The most important finding was observing what experiences were independently related to profound FT. Participants with higher FT attempted to raise funds for treatment expenses by cutting through their deposits and savings, while reducing spending on basics like food or clothing. This finding can expand on previous findings regarding the relationship between FT and the use of financial coping strategies [9]. It would be important for oncology professionals to assess how financial burden force patients to change their lives as above. Of note, any financial status was not significantly associated with FT in multiple regression analysis. Participants might have to worry about future payment even if they had more income and/or less expenditure. Worry about financial problems differ from actual financial problems, such as inability to pay healthcare costs [20]. Our findings suggest that FT in this population could heighten due to financial worry rather than due to a direct impact of financial burden. Worry about affording healthcare predicts more cost-related non-adherence [21]. Similarly, in this study, financial hardship put participants at a crossroads regarding whether they should continue cancer treatment, thus supporting the importance of intervention to FT caused by molecular-targeted and immune therapy.

Our results also suggest that FT might become severe as the impact of financial burden on patient experiences spread, and vice versa. Participants with higher FT reduced spending on basics like food or clothing, and further made up for shortage of funds for treatment expenses by cutting through their deposits and savings, which resulted in hesitating continuing treatment based on finances. In fact, participants’ FT was significantly related to the number of patient experiences associated with the financial burden. Patients would experience profound FT as a result of the accumulation of diverse experiences associated with financial burden. This finding is important in that it confirms that multifaceted aspects of experience associated with the financial burden of cancer treatment can cause FT. Material conditions (financial spending and financial resources), psychosocial responses (affect), and coping behaviors (support seeking, coping care, and coping lifestyle) have been proposed as conceptual domains and subdomains for subjective financial distress in patients with cancer [3]. Our results are consistent with the conceptual framework and suggest that participants experience profound FT not only from the material use of their deposits and savings but also from psychosocial and behavioral coping, owing to continuous payments for treatment. Oncology professionals must have a multifaceted understanding of the financial burden experienced by patients.

This study had some limitations. First, this study was an observational study without control (e.g., cancer patients who received cytotoxic chemotherapy). We could not refer to characteristics of FT in molecular-targeted and immune therapy in comparison with that in other cancer treatments. Also, we recruited participants irrespective of the price of the drug that they used. Second, the participants were recruited by convenience sampling and were small in number. Selection bias cannot be thus avoided. Moreover, the age restriction of the inclusion criteria limits the generalizability of the findings. Third, this was a cross-sectional study, hence we could not identify the causal relationship between FT and patient experience. The reverse possibility thus needs to be considered. Fourth, the participants’ responses to financial burden might have been influenced by recall bias because they were asked to respond to the period since the start of the cancer treatment they were receiving. The participants evaluated the experience for a median of 11 months. Finally, we could not consider the impact of other factors regarding disease and treatment. Notably, physical conditions, such as types of cancer, disease status, and symptoms related to cancer and its treatment, need to be considered because the presence of comorbidities was a significant factor of FT. Despite these limitations, the findings of this study revealed the importance of focusing on patient experiences to consider the financial aspect of molecular-targeted and immune therapies. Multicenter recruitment and a high response rate are the strengths of this study.

In conclusion, patients receiving molecular-targeted and immune therapies for cancer are at risk of experiencing profound FT, leading to a reduced quality of life. Profound FT can be caused by multifaceted experiences associated with the financial burden of cancer treatment. The financial burden of continuing molecular-targeted and immune therapies can even make patients hesitant to continue cancer treatment. Oncology professionals, including physicians, nurses, pharmacists, medical affairs, and social workers, need to consider multifaceted aspects of patient experience linked to financial burden and screen financial worry before actual financial problems appear. The independently related factors that we identified have the potential to serve as indicators of profound FT and the need for specialized intervention.