Introduction
Multi-institutional electronic medical record (EMR) databases, such as the Medical Information Database Network (MID-NET), have been recommended for postmarketing studies in Japan. However, their utility for cancer pharmacoepidemiology studies is not well-established.

Aims
To assess the validity of oncology-related data in EMR databases compared to those in hospital cancer registries (HCR).

Methods
Using EMR and HCR data (2011-2021) from two institutions: Site-A (a large university/teaching hospital) and Site-B (10 community hospitals from a large national healthcare system), both contributing to MID-NET, we identified incident cases for 11 cancer types using “diagnosis category” and the “initial/recurrent” variable in EMR data. We calculated the positive predictive value (PPV) and sensitivity of incident cancer in EMR using HCR as the gold-standard. Initial cancer treatments and stages were compared between EMR and HCR. We also examined the length of the available baseline period for covariate/confounder ascertainment.

Results
Among 102,735 to 472,078 patients with one of 11 cancer types in EMR, 14% (14,164) and 6% (28,542) had incident cancer in Site-A and Site-B, respectively. The PPV for EMR-based incident cancer was 85% and 83%, and the sensitivity was 79% and 68% in Site-A and Site-B, respectively. Cancer treatments and stages recorded in EMR matched HCR records exactly. Patients with baseline data available for over 180 days prior to diagnosis accounted for only 35% and 48% of all incident cases in Site-A and Site-B, respectively. Varying baseline periods from 30 to 180 days, approximately 50% of covariates captured using the 180-day period were identified using the 30-day period.

Discussion/conclusion
While 20-30% of incident cases were missed and ~15% of EMR-base incident cases were not incident in EMR, initial treatments/stages were identified with 100% accuracy. Short baseline periods are concerning for confounding adjustment, potentially missing 50% of baseline covariates using a 30-day baseline period.

Introduction: Northern India ranks second highest globally for gallbladder cancer (GBC) cases, and there’s a need for more regional studies on the link between arsenic in drinking water and GBC due to significant geographical variations.
Aim: To explore the risk and potential carcinogenic role of arsenic and other heavy metals in the drinking water of gallbladder cancer patients.
Methodology: We conducted a hospital-based observational cohort study in a tertiary care cancer specialty hospital in northeast India. Using inductively coupled mass spectrometry (ICPMS), an Agilent 7800, arsenic (As), and other 11 trace elements (Be, V, Cr, Fe, Ni, Co, Se, Sr, Cd, Hg, and Pb) were measured in urine (n=216) and drinking water (n=97) collected from histopathologically confirmed GBC patients. The USEPA model was used to assess cancer and non-cancer risks. The Incremental Lifetime Cancer Risk (CR) and Hazard Quotient (HQ) were used to determine cancerous and non-cancerous risk levels in patients with GBC. Heavy metal concentrations were expressed in parts per billion ± standard error.
Results: The mean concentration of As (20.59±2.09), Se (78.49±17.95), Sr (82.05±7.08), and Cd (4.86±0.65) exceeded the WHO limit in urine samples of GBC patients. About 61.3% of water samples exceeded the WHO limit for As in their drinking water. The health risk assessment shows that the exposed concentration of As, Cr, Ni, and Cd in the urine samples can cause cancer risk with CR values of 0.001, 0.001, 0.000089, and 0.0012, respectively. The As, Cr, Ni, and Cd (CR: 0.00007, 0.0003, 0.00044, and 0.00001, respectively) had cancer risk with the exposed concentration in their drinking water. There was a strong positive correlation in arsenic concentrations between water and urine (R2=0.84).
Conclusion: The findings reveal a significant association between arsenic, heavy metal contamination, and gallbladder carcinogenesis, warranting further investigation through individual-level observational and molecular studies.

Introduction:
Epidermal growth factor receptor (EGFR) exon 20 insertion (exon20ins) is an uncommon mutation in non-small cell lung cancer (NSCLC) patients, which had poor prognosis outcome prior to the introduction of novel target agents (i.e., amivantamab). Treatment options for this population remain limited.

Aims:
To analyze treatment pattern and clinical outcomes in NSCLC patients with EGFR exon20ins.

Methods:
Electronic medical record data from Chang Gung Research Database, a multi-institutional network containing around 10% of Taiwanese patients, were extracted and transformed into a common data model specially for the EXPLORE-LC Taiwan dataset. Biomarker data were further enhanced. All newly diagnosed stage IIIB/IV NSCLC patients with EGFR exon20ins who aged 18 years or older and received first NSCLC treatment between 2016 to 2020 were included and followed up until death or December 31 2022. Patient characteristics were described descriptively. Treatment sequences and clinical outcomes were analyzed and stratified by different types of first-line systematic anti-cancer treatments (SACT). Overall survival (OS) was estimated using the Kaplan-Meier method with 95% confidence interval (CI) reported.

Results:
Among 4,445 stage IIIB/IV NSCLC patients included in the dataset, only 181 patients (4.1%) had EGFR exon20ins. Of these EGFR exon20ins mutated patients, 170 (93.9%) were non-squamous carcinoma and 42 (23.2%) had brain metastasis at diagnosis. Furthermore, eight (4.4%) and four (2.2%) patients co-mutated with ALK and ROS1, respectively. During the follow-up period, 175 (96.7%) patients received at least one line of SACT. EGFR tyrosine kinase inhibitors (EGFR-TKI; n=121, 69.1%) and platinum-based chemotherapy (n=39, 22.3%) were two main first-line SACT regimens. Patients treated with EGFR-TKI had better median OS than those receiving platinum-based chemotherapy (32.3 [95% CI: 24.9, 46.5] vs. 15.9 [95% CI: 10.0, 22.4] months).

Discussion:
The study provides additional data on the real-world characteristics, treatment patterns, and clinical outcomes of Taiwanese NSCLC patients with EGFR exon20ins.

Introduction: Whether concomitant use of angiotensin converting enzyme inhibitors (ACEi) and dipeptidyl peptidase 4 inhibitors (DPP4i), both potentiating the levels of bradykinin and substance P, would further exaggerate the risk of lung cancer remains uncertain.
Aims: To evaluate whether the concomitant use of ACEi and DPP4i is associated with an increased risk of lung cancer compared with concomitant use of ACEi and SGLT2i.
Methods: Using the nationwide healthcare data of South Korea between 2010 and 2022, we identified a cohort of adults who firstly received ACEi and then identified a subset of patients firstly prescribed DPP4i or SGLT2 inhibitors (SGLT2i) during the ACEi treatment. The index date was defined as the first date of DPP4 inhibitors or SGLT2i administration. The primary outcome was incident lung cancer, defined using a previously validated algorithm with an 88.9% positive predictive value. Patients were followed from 1 year after the index date until the earliest occurrence of lung cancer, 1 year after switching between the study drug classes, death, or study end date (31 December 2022). Hazard ratios (HRs) and rate differences (RDs) per 1000 person-years of lung cancer were estimated after weighting by propensity score fine stratification method.
Results: Among a total of 42,108 eligible patients initiating ACEi, 35,990 (85.5%) initiated DPP4i and 6,118 (14.5%) initiated SGLT2i. During a 5.2-year mean follow-up, concomitant use of DPP4i vs. SGLT2i with ACEi was associated with an elevated risk of lung cancer (2.28 vs. 1.18 events per 1000 person years; HR 1.95, 95% CI 1.09 to 3.49; RD 1.10, 0.50 to 1.70).
Discussion/Conclusion: Among adults initiating ACEi, concomitant use of DPP4i was associated with an increased risk of incident lung cancer compared with SGLT2i. Thus, concomitant use of ACEi and DPP4i poses a significant drug-drug interaction, and this combination of therapy should be used with caution.

Introduction:
The state-of-the-art of the treatment for head and neck cancer (HNC) has evolved recently due to the development of immunotherapy and advances in radiotherapy. However, the treatment modalities applied in real-world settings in Japan are not fully described.
Aims: To investigate the initial treatment modalities of primary HNC in real-world settings in Japan.
Methods:
A retrospective cohort study using the Japan Medical Data Center (JMDC) was conducted. Adult patients with at least two diagnosis codes of the same primary HNC cancer (ICD-10-CM: C00-C14, C32) between January 2015 and March 2023 initiating HNC treatment were included. Patients were followed till the end of initial treatment, last enrollment in JMDC, or June 30th, 2023, whichever happened first. The initial treatment episode was defined as the period with the first HNC treatment before a 90-day gap. The cohort was classified into four groups based on surgery and metastasis status: resectable, unresectable without mention of metastasis (unresect-NM), unresectable with regional metastasis (unresect-RM), and unresectable with distant metastasis (unresect-DM). The treatment patterns were described for each group.
Results:
A total of 3,031 patients were included in the study with 41.1%, 36.2%, 16.5% and 6.2% in resectable, unresect-NM, unresect-RM, and unresect-DM group, respectively. Chemoradiotherapy was the most common treatment modalities among unresectable patients (unresect-NM: 54%; unresect-RM: 76%; unresect-DM: 58%), and was the second common treatment modalities among resectable patients (20%) following surgery only (62%). 29% of patients in unresect-DM group used PD-(L)1 inhibitor, however, only 3%-8% of the other patients received it. Among 1,960 patients receiving radiotherapy (resectable: 20.4%; unresect-NM: 48.5%; unresect-RM: 23.9%; unresect-DM: 7.2%), only 30 patients received advanced particle therapies such as proton therapy, heavy ion therapy, and boron neutron capture therapy.
Discussion/Conclusion:
One-third of the patients in unresect-DM group received immunotherapy, but the adoption of advanced particle therapies remains low.

Introduction. Antineoplastic agents commonly cause adverse drug reactions (ADRs) in cancer patients, who are particularly vulnerable due to low tolerance. Pharmacovigilance of these drugs is limited due to significant under-reporting, making it crucial to measure the frequency and severity of ADRs in oncology patients.
Aim. We aimed to assess the frequency, types, severity, and management of ADRs in chemotherapy patients.
Methods. A 12-month cross-sectional study was conducted in a tertiary-care teaching hospital, including 327 chemotherapy patients from January to December 2023. The data was collected in a suspected ADR reporting form by the Central Drugs Standard Control Organization (CDSCO) and reported to the ADR Monitoring Centre and Vigiflow at the National Coordinating Centre (NCC). The suspected drugs were evaluated for causality using the Modified Naranjo scale, and ADR severity was assessed with the Hartwig et al. severity scale.
Results. Out of 327 patients, 230 developed ADR (70%). A total of 372 ADRs were reported, which indicated 1.6 events per patient. ADRs mainly occurred in the age group of 40-60 years (57.0%), predominantly affecting females (64.8%). Common ADRs included thrombocytopenia (11.2%), neutropenia (9.9%), and diarrhea (7.7%). Ovarian (17.3%) and breast cancer patients (15.6%) had the highest ADR incidence. Platinum-based agents and antimetabolites were the common culprit drugs. Causality assessment showed 36.1% of ADRs as ‘possible’ and 61.7% as ‘probable.’ Severity of ADRs was categorized as ‘mild’ (49.13%), ‘moderate’ (46.08%), and ‘severe’ (1.3%). All patients received corrective measures upon experiencing ADR. In addition, chemotherapy was temporarily withheld in 49 patients until the improvement with corrective measures. The therapeutic agent was substituted in two patients, and the dosage was reduced in eight patients.
Conclusion. The study emphasizes the critical need for vigilant surveillance to minimize the impact of these complications on treatment outcomes. Future studies could stratify data based on cancer stages. Developing a model for early detection of severe ADRs should also be explored to improve patient outcomes.