Takahiro Higashi

Introduction: Thymomas and thymic carcinomas are rare diseases with limited clinical evidence. This study aimed to investigate the clinical characteristics of and treatment practices for Stages III and IV thymoma or thymic carcinoma in Japan. Methods: We analyzed cases of Stage III and IV thymoma and thymic carcinoma diagnosed between 2018 and 2022 and recorded in a nationwide dataset linking the Hospital‐based Cancer Registry and the Diagnosis Procedure Combination survey. Treatment practices, including the use of perioperative chemotherapy, radiotherapy, and newer systemic therapies as well as the chemotherapy regimens, were evaluated. Results: Overall, 556 stage III thymoma, 312 stage III thymic carcinoma, 431 stage IV thymoma, and 976 stage IV thymic carcinoma patients were identified. Postoperative radiotherapy after complete resection was more common for stage III thymic carcinoma than for thymoma (34.1% vs. 10.7%), while perioperative chemotherapy was used more frequently for stage IV thymoma than for thymic carcinoma (preoperative: 15.5% vs. 3.7%; postoperative: 8.4% vs. 4.4%). For stage IV thymoma, adriamycin–cisplatin–vincristine–cyclophosphamide was the most common regimen (41.8%), whereas carboplatin–paclitaxel was predominantly used for stage IV thymic carcinoma (99.5%). The proportion of anthracycline‐based therapy for stage IV thymoma increased from 50.0% to 77.8%, while that of postoperative radiotherapy for stage III thymoma decreased from 11.1% to 4.1%. Conclusions: Evidence‐based treatment approaches for thymoma and thymic carcinoma have been widely adopted in Japan.

Purpose Financial toxicity (FT) refers to increased risk of treatment discontinuation and anxiety in daily life due to the cost burden of cancer treatment. However, little is known about which groups of patients are more likely to experience FT under universal health coverage. Thus, we aimed to explore demographics and psychological factors associated with FT among adult cancer patients in Japan. Methods We analyzed responses from the Patient Experience Survey, a self-administered questionnaire conducted by the National Cancer Center in 2018. Outcome indicators were aligned with the Phase 4 Basic Plan to Promote Cancer Control Programs in Japan. We used a modified Poisson regression model to analyze the association between patient demographics and FT, estimating the Prevalence Ratio (PR) and conducting subgroup analysis by sex. Results Of 6766 responses received, approximately 30% of patients reported FT. FT was categorized into “FT impacted on daily life (FT_Life),” “FT impacted on treatment-receiving behavior (FT_Tx),” and “FT led to treatment change or discontinuation (FT_DisTx).” Across all FT categories, younger age (< 40 vs. 65 + ; male: PR = 2.26 [95% CI, 0.98–5.23]; female: PR = 1.79 [95% CI, 1.01–3.17]), and those with no-one to talk to (vs. with; male: PR = 2.04 [95% CI, 1.05–3.94]; female: PR = 2.07 [95% CI, 1.09–3.93]) were consistently associated with FT. Conclusion These results identify groups that require financial assistance due to cancer treatment, including younger patients, individuals with unstable employment, and those with limited interpersonal support. Strengthening social support systems is critical to enable these patients to manage treatment costs and maintain daily life.

Introduction Anticancer therapy for non-small cell lung cancer in patients ≥ 75 years has evolved with accumulating evidence. Recent guidelines recommend that treatment decisions be based on functional status rather than age alone. However, disparities in treatment rates persist between those above and below 75 years. This study examined whether evidence supporting treatment in older patients influences implementation, comparing trends over time by age groups. Methods A hospital-based cancer registry covering designated cancer centers was linked with a Diagnosis Procedure Combination database, which records all medical interventions at each facility, capturing approximately 50% of lung cancer cases in Japan. For patients ≥ 75 years with good performance status, the implementation rates of standard anticancer therapy were analyzed by disease stage in 2021 and trends from 2016 to 2021. Results In 2021, all standard treatments, except oral molecular-targeted therapies for patients with unresectable cancer, were administered less frequently to those ≥ 75 years. Treatments with evidence established before 2016 were stable throughout the study period. Oral molecular-targeted therapies demonstrated no age-related differences, and adjuvant cytotoxic chemotherapy was less frequently administered to patients ≥ 75 years. In contrast, treatments validated in patients ≥ 75 years after 2016 demonstrated a significant trend towards higher implementation, approaching that in younger patients. Immune checkpoint inhibitors, for which safety and efficacy in older patients remain uncertain, exhibited increasing divergence in implementation rates between age groups over time. Conclusion Evidence supporting the efficacy and safety of therapies in individuals ≥ 75 years may influence treatment selection over time.

Purpose In this study, we assessed patient experiences with cancer care in Japan as part of the national evaluation of the National Cancer Control Plan. We aimed to identify strengths and gaps in cancer care delivery across patient groups, including those with rare cancers and individuals aged < 40 years, to inform future policy development. Methods We conducted a nationwide cross-sectional survey between December 2023 and March 2024 using a self-administered mail questionnaire. Eligible participants were selected from the national database of Hospital-Based Cancer Registries using a two-stage stratified random sampling method. Among the 32,716 individuals invited, 11,247 eligible responses were analyzed. Sampling weights were applied to ensure national representativeness. Outcomes included overall ratings of care, information provision, access to care, symptom burden, psychosocial support, and socioeconomic issues. Results Patients rated the care they received highly (mean score = 8.2/10); however, notable gaps were identified. Only 55.0% of respondents were aware of the Cancer Counseling and Support Centers, and merely 15.4% were informed about peer support. Patients with rare cancers and those aged < 40 years faced delayed diagnosis, greater difficulty accessing care, and heavier psychosocial and financial burdens. Younger patients were more likely to report emotional distress and unmet informational needs regarding work-related issues than older patients. Conclusion Cancer care in Japan is generally well-received; however, significant challenges remain, particularly for younger patients and those with rare cancers. Increasing support service awareness and utilization, along with improved access pathways and targeted support for adolescent and young adult patients, are essential to promoting equitable and patient-centered cancer care under the revised Cancer Control Plan.

Background: Generative artificial intelligence (AI) is increasingly used to find information. Providing accurate information is essential to support patients with cancer and their families; however, information returned by generative AIs is sometimes wrong. Returning wrong information is called hallucination. Retrieval-augmented generation (RAG), which supplements large language model (LLM) outputs with relevant external sources, has the potential to reduce hallucinations. Although RAG has been proposed as a promising technique, its real-world performance in public health communication remains underexplored. Objective: This study aimed to examine cancer information returned by generative AIs with RAG using cancer-specific information sources and general internet searches to determine whether using RAG with reliable information sources reduces the hallucination rates of generative AI chatbots. Methods: We developed 6 types of chatbots by combining 3 patterns of reference information with 2 versions of LLMs. Thus, GPT-4 and GPT-3.5 chatbots that use cancer information service (CIS) information, Google information, and no reference information (conventional chatbots) were developed. A total of 62 cancer-related questions in Japanese were compiled from public sources. All responses were generated automatically and independently reviewed by 2 experienced clinicians. The reviewers assessed the presence of hallucinations, defined as medically harmful or misinformation. We compared hallucination rates across chatbot types and calculated odds ratios (OR) using generalized linear mixed-effects models. Subgroup analyses were also performed based on whether questions were covered by CIS content. Results: For the chatbots that used information from CIS, the hallucination rates were 0% for GPT-4 and 6% for GPT-3.5, whereas those for chatbots that used information from Google were 6% and 10% for GPT-4 and GPT-3.5, respectively. For questions on information that is not issued by CIS, the hallucination rates for Google-based chatbots were 19% for GPT-4 and 35% for GPT-3.5. The hallucination rates for conventional chatbots were approximately 40%. Using reference data from Google searches generated more hallucinations than using CIS data, with an OR of 9.4 (95% CI 1.2‐17.5, P<.01); the OR for the conventional chatbot was 16.1 (95% CI 3.7‐50.0, P<.001). While conventional chatbots always generated a response, the RAG-based chatbots sometimes declined to answer when information was lacking. The conventional chatbots responded to all questions, but the response rate decreased (36% to 81%) for RAG-based chatbots. For questions on information not covered by CIS, the CIS chatbots did not respond, while the Google chatbots generated responses in 52% of the cases for GPT-4 and 71% for GPT-3.5. Conclusions: Using RAG with reliable information sources significantly reduces the hallucination rate of generative AI chatbots and increases the ability to admit lack of information, making them more suitable for general use, where users need to be provided with accurate information.

Supplementary Fig. S7. Treatment patterns stratified by age group and sex in patients with distant metastasis.