From February 15 to 19, 2024, a total of 1,261 participants were recruited for the nationwide survey. Participants characteristics are summarized in Table 1. The median age was 63 years for CA participants, 54 for MP participants, and 59 for non-CA participants. MP exhibited a younger median age which is attributable to the predominance of working-age individuals (≤ 60 years) among participants. There were no notable differences among the participants in terms of marital status or number of children in the household.

Table 1 Characteristics of survey participants

The medical history of survey participants is presented in Supplementary Fig. S2A. Among the MP group, approximately half were medical doctors (49%), followed by nurses (36%), pharmacists (12%), and clinical laboratory technicians (3%). Among the CA group, the three most common cancer types were breast cancer (N = 136), gastrointestinal/hepatobiliary/pancreas cancer (N = 127), and urinary cancer (N = 88, Supplementary Fig. S2B). Notably, 35 participants (8%) in the CA group also reported working as medical professionals.

Comprehension levels of explanatory materials: Responses to “How well did you understand the explanatory materials about MBRT?”

Across the total population, more than 60% of participants demonstrated good comprehension of each explanatory material as assessed by Top-2 box responses on a 5-point Likert scale (Table 2). For “Options of cancer treatment,” 71% of CA participants, 81% of MP participants, and 50% of non-CA participants selected Top-2 box responses (Excellent or Very well). For “Anticancer treatments based on the cancer genomic alteration,” the corresponding proportion in Top-2 box were 57% of CA participants, 80% of MP participants, and 44% of non-CA participants, respectively. For “The Japanese healthcare system and off-label drug use,” 57% of CA participants, 77% of MP participants, and 47% of non-CA participants indicated high levels of understanding. For “Notes on treatment with the off-label drug use,” Top-2 box responses were 60% of CA participants, 78% of MP participants, and 49% of non-CA participants.

Table 2 Comprehension levels of explanatory materials

Overall comprehension of the 4-page explanatory material was “GOOD” for 651 participants, “FAIR” for 288 participants, and “POOR” for 322 participants.

Willingness to utilize off-label drugs: Responses to “Would you consider utilizing the off-label anticancer drug as MBRT identified by CGP testing?”

Participants were asked about their willingness to use off-label anticancer drugs in three different cost conditions: (1) no cost burden, (2) ¥200,000 per month (approximately $US1,300), and (3) ¥1,000,000 per month (approximately $US6,700) drug cost burden. Responses in the Top-2 box on a 5-point Likert scale (Very positive or Positive) were categorized as willing to use.

The results showed that approximately half or more of the participants in all groups expressed willingness to use off-label drugs when there was no cost burden (Fig. 1A and Table 3). This was most prominent among MP participants (62%, 95% confidence interval [CI] 57–67%), compared with 51% (95% CI 46–56%) of CA participants and 50% (95% CI 45–55%) of non-CA participants.

Fig. 1

The percentage of participants who would like to use the off-label drug according to cost burden. A, Vertical bar graphs show the percentage of those willing to use the off-label drug—cancer patients (dark gray), medical professionals (light gray), and non-cancer volunteers (gray)—according to the cost burden per month (zero, ¥200,000, and ¥1,000,000). *p < 0.001 vs no cost burden in cancer patients. †p < 0.001 vs no cost burden in medical professionals. ‡p < 0.001 vs no cost burden in non-cancer volunteers. B, Willingness to utilize off-label drugs according to overall comprehension of explanatory materials. Vertical bar graphs show the percentage of individuals willing to use the off-label drug in GOOD overall comprehension (dark gray), FAIR overall comprehension (light gray), and POOR overall comprehension (gray) according to the cost burden per month (zero, ¥200,000, and ¥1,000,000). Error bars (A, B) represent the 95% confidence interval.

Table 3 Willingness to utilize the off-label drug as molecular-based recommended therapy identified by comprehensive genomic profiling

In contrast, the proportion of participants willing to use the off-label drug declined significantly as the cost burden increased, with similar trends across all participant groups (Fig. 1A and Table 3). When the cost was ¥200,000 per month, 15% (95% CI 12–19%) of CA participants, 31% (27–36%) of MP participants, and 15% (12–19%) of non-CA were willing to use the off-label drug (p < 0.001 vs no cost burden in each participant group). When the cost increased to ¥1,000,000 per month, willingness further declined to 6% (4–9%) of CA participants, 16% (13–20%) of MP participants, and 5% (3–7%) of non-CA participants (p < 0.001 vs no cost burden in each participant group).

The proportion of participants willing to use the off-label drug with no cost burden was similar between men and women across all groups (Supplementary Table S1). However, when the cost exceeded ¥200,000 per month, the proportion of women who were willing to use the drug decreased more markedly than that of men. A decline in willingness to use the off-label drug with an increasing cost burden was observed across all ages of participants, indicating that this trend was independent of age.

Participants were also asked to provide reasons for their responses to the question on willingness to use off-label anticancer drugs. The most common reason among those who selected Top-2 box responses (Very positive or Positive) was “I expect the anti-tumor efficacy,” cited by 60–75% of respondents (Table 4). This was followed by “I don’t want to do nothing” and “I believe I can contribute to future cancer treatment.” These trends were consistent across all cost burden options. While many MP expressed willingness to pay higher costs in expectation of anti-tumor efficacy, CA and non-CA showed low willingness to do so (Supplementary Table S2). Among participants who responded Fair, Negative, or Not at all, the most frequently cited reason was cost, followed by “uncertain efficacy,” “concerns about safety,” “cancer is incurable,” and “psychological resistance to off-label drug drugs” (Table 4). These trends were also consistent across all cost burden conditions in each participant group (Supplementary Table S2).

Table 4 Reasons for answers about willingness to utilize the off-label drug as molecular-based recommended therapyWillingness to utilize off-label drugs according to overall comprehension of explanatory materials

The level of overall comprehension of the explanatory materials demonstrated a significant positive correlation with the proportion of individuals willing to use off-label anticancer drugs (Fig. 1B). In the no-cost burden condition, willingness to use off-label drugs was significantly higher in the GOOD comprehensive participants (65%, 95% CI 61–69%) compared with the “FAIR” participants (57%, 95% CI 51–62%, p < 0.05) and the “POOR” participants (31%, 95% CI 23–37%, p < 0.001). This trend was consistently observed across both the no-cost burden and higher-cost burden conditions, with a statistically significant increase in willingness to use off-label drugs correlated with higher levels of comprehension of the provided information (Fig. 1B). Similar trends were also observed within each participant group: CA, MP, and non-CA (Supplementary Fig. S3A, B, C).