Impact of deep learning-determined smoking status on mortality of cancer patients: never too late to quit

Applications of Natural Language Processing in Otolaryngology: A Scoping Review

OBJECTIVE

To review the current literature on the applications of natural language processing (NLP) within the field of otolaryngology.

DATA SOURCES

MEDLINE, EMBASE, SCOPUS, Cochrane Library, Web of Science, and CINAHL.

METHODS

The preferred reporting Items for systematic reviews and meta-analyzes extension for scoping reviews checklist was followed. Databases were searched from the date of inception up to Dec 26, 2023. Original articles on the application of language-based models to otolaryngology patient care and research, regardless of publication date, were included. The studies were classified under the 2011 Oxford CEBM levels of evidence.

RESULTS

One-hundred sixty-six papers with a median publication year of 2024 (range 1982, 2024) were included. Sixty-one percent (102/166) of studies used ChatGPT and were published in 2023 or 2024. Sixty studies used NLP for clinical education and decision support, 42 for patient education, 14 for electronic medical record improvement, 5 for triaging, 4 for trainee education, 4 for patient monitoring, 3 for telemedicine, and 1 for medical translation. For research, 37 studies used NLP for extraction, classification, or analysis of data, 17 for thematic analysis, 5 for evaluating scientific reporting, and 4 for manuscript preparation.

CONCLUSION

The role of NLP in otolaryngology is evolving, with ChatGPT passing OHNS board simulations, though its clinical application requires improvement. NLP shows potential in patient education and post-treatment monitoring. NLP is effective at extracting data from unstructured or large data sets. There is limited research on NLP in trainee education and administrative tasks. Guidelines for NLP use in research are critical.

Large language models in cancer: potentials, risks, and safeguards

This review examines the use of large language models (LLMs) in cancer, analysing articles sourced from PubMed, Embase, and Ovid Medline, published between 2017 and 2024. Our search strategy included terms related to LLMs, cancer research, risks, safeguards, and ethical issues, focusing on studies that utilized text-based data. 59 articles were included in the review, categorized into 3 segments: quantitative studies on LLMs, chatbot-focused studies, and qualitative discussions on LLMs on cancer. Quantitative studies highlight LLMs' advanced capabilities in natural language processing (NLP), while chatbot-focused articles demonstrate their potential in clinical support and data management. Qualitative research underscores the broader implications of LLMs, including the risks and ethical considerations. Our findings suggest that LLMs, notably ChatGPT, have potential in data analysis, patient interaction, and personalized treatment in cancer care. However, the review identifies critical risks, including data biases and ethical challenges. We emphasize the need for regulatory oversight, targeted model development, and continuous evaluation. In conclusion, integrating LLMs in cancer research offers promising prospects but necessitates a balanced approach focusing on accuracy, ethical integrity, and data privacy. This review underscores the need for further study, encouraging responsible exploration and application of artificial intelligence in oncology.

Unveiling the Influence of AI on Advancements in Respiratory Care: Narrative Review

BACKGROUND

Artificial intelligence is experiencing rapid growth, with continual innovation and advancements in the health care field.

OBJECTIVE

This study aims to evaluate the application of artificial intelligence technologies across various domains of respiratory care.

METHODS

We conducted a narrative review to examine the latest advancements in the use of artificial intelligence in the field of respiratory care. The search was independently conducted by respiratory care experts, each focusing on their respective scope of practice and area of interest.

RESULTS

This review illuminates the diverse applications of artificial intelligence, highlighting its use in areas associated with respiratory care. Artificial intelligence is harnessed across various areas in this field, including pulmonary diagnostics, respiratory care research, critical care or mechanical ventilation, pulmonary rehabilitation, telehealth, public health or health promotion, sleep clinics, home care, smoking or vaping behavior, and neonates and pediatrics. With its multifaceted utility, artificial intelligence can enhance the field of respiratory care, potentially leading to superior health outcomes for individuals under this extensive umbrella.

CONCLUSIONS

As artificial intelligence advances, elevating academic standards in the respiratory care profession becomes imperative, allowing practitioners to contribute to research and understand artificial intelligence's impact on respiratory care. The permanent integration of artificial intelligence into respiratory care creates the need for respiratory therapists to positively influence its progression. By participating in artificial intelligence development, respiratory therapists can augment their clinical capabilities, knowledge, and patient outcomes.

The association between breast cancer and lung cancer: a bidirectional Mendelian randomization study

With increasing life spans, breast cancer (BC) survivors may face the possibility of developing second primary cancer (SPC), which can considerably shorten survival. Lung cancer (LC) is a common SPC among BC survivors. This study explored the association between these two cancers through Mendelian randomization (MR) analysis. A bidirectional two-sample MR analysis was conducted with BC genome-wide association study (GWAS) data from the Breast Cancer Association Consortium (BCAC) included 228,951 individuals and the GWAS summary statistics from the Transdisciplinary Research in Cancer of the Lung (TRICL) of LC included 112,781 individuals. The IVW method and MR-RAPS method showed a causal effect of overall BC on lung adenocarcinoma (LUAD) (IVW: OR = 1.060, 95% CI = 1.008-1.116, P = 0.024; MR-RAPS: OR = 1.059, 95% CI = 1.005-1.116, P = 0.033), which indicated that patients with BC had an increased risk of LUAD. However, there is no strong evidence for a causal effect of LUAD on BC. Our study revealed a causal effect of BC on second primary LUAD, suggesting that we should intensify screening for second primary LC in BC survivors. Early intervention and treatment for patients with second primary LC are needed to reduce mortality in BC survivors.

Effect of smoking on melanoma incidence: a systematic review with meta-analysis

BACKGROUND

There is a strong correlation between cigarette smoking and the development of many cancer types. It is therefore paradoxical that multiple reports have suggested a reduced incidence of melanoma in smokers. This study aimed to analyze all existing studies of melanoma incidence in smokers relative to nonsmokers.

METHODS

Searches of MEDLINE and Embase were conducted for studies reporting data on melanoma in smokers and never-smokers. No study design limitations or language restrictions were applied. The outcome examined was the association between smoking status and melanoma. Analyses focused on risk of melanoma in smokers and never-smokers generated from multivariable analyses, and these analyses were pooled using a fixed-effects model. Risk of bias was assessed using the Newcastle-Ottawa tool.

RESULTS

Forty-nine studies that included 59 429 patients with melanoma were identified. Pooled analyses showed statistically significant reduced risks of melanoma in male smokers (risk ratio [RR] = 0.60, 95% confidence interval [CI] = 0.56 to 0.65, P < .001) and female smokers (RR = 0.79, 95% CI = 0.73 to 0.86, P < .001). Male former smokers had a 16% reduction in melanoma risk compared with male never-smokers (RR = 0.84, 95% CI = 0.77 to 0.93, P < .001), but no risk reduction was observed in female former smokers (RR = 1.0, 95% CI = 0.92 to 1.08).

CONCLUSIONS

Current smokers have a statistically significant reduced risk of developing melanoma compared with never-smokers, with a reduction in melanoma risk of 40% in men and 21% in women.

Outcomes After Multimodality Treatment of Pancreatic Cancer in an Unselected Single-Center Cohort

Background

Pancreatic ductal adenocarcinoma (PDAC) remains a lethal and rarely resectable malignancy. Here we explore the outcomes of surgery, as compared to definitive radiotherapy (dRT) or systemic therapy only in PDAC.

Methods

Pancreatic surgery and radiotherapy in Southwest Finland have been centralized to Turku University Hospital. Previously validated population-based electronic health records database was searched for all unselected PDAC patients from the years 2009-2019. Main outcome was median overall survival (mOS). Demographics, pathology, surgery, and oncological treatment data were collected.

Results

We identified 1006 patients with PDAC, 49% male, median age 71 years and 77% presenting with metastatic disease. In total, 405 patients were treated; 92 resected, 26 dRT without resection and 287 systemic therapy only. mOS was 34.6 months for resected, 26.7 months for dRT, and 7.5 months for systemic therapy patients. Among the 88 patients with locally advanced inoperable PDAC, dRT was independently associated with longer mOS (26.7 months) as compared to systemic therapy only (mOS 10.6 months). Among the 287 patients treated with systemic therapy only, combination chemotherapy was independently associated with longer mOS (11.6 months) as compared to gemcitabine-monotherapy (6.8 months). In patients progressing to second-line systemic treatment after gemcitabine failure, mOS was the same (5.0 months) with single or combination regimens.

Conclusion

Surgery remains the only curative approach for PDAC. In locally advanced PDAC, dRT was associated with longer survival as compared to systemic therapy only. Concerning first-line systemic therapy, our results support the use of combination chemotherapy over single-agent therapy.

The impact of prognostic factors and comorbidities on survival in older adults with stage I - III cutaneous melanoma in Southwest Finland: A register study

INTRODUCTION

Despite being diagnosed with thicker and more often ulcerated melanomas, cancer-specific survival (CSS) is not necessarily inferior in older adults with melanoma compared to younger patients.

MATERIALS AND METHODS

Our aim was to evaluate the impact of baseline melanoma-specific prognostic factors and comorbidities on recurrence-free survival (RFS), CSS, and overall survival (OS) in patients aged 70-79 (n = 474) and ≥ 80 years (n = 286) with resected stage I - III cutaneous melanoma in Southwest Finland between January 1, 2000 and December 31, 2020. Patients were restaged according to the 8th edition of TNM classification, and comorbidities were assessed using the Charlson Comorbidity Index (CCI).

RESULTS

Patients aged ≥80 years had thicker and more commonly ulcerated melanomas: 43.0%, 40.9%, and 16.1% of patients aged ≥80 and 56.5%, 25.3%, and 18.1% of patients aged 70-79 years were diagnosed with stage I, II, and III melanoma, respectively. Multiple comorbidities (CCI ≥2) were more common and sentinel lymph node biopsy less frequently performed in patients aged ≥80 years. RFS and CSS were similar in patients aged 70-79 years and ≥ 80 years: median RFS 13.8 years vs not reached, with the hazard ratio of melanoma recurrence or death from melanoma 1.25 (95% confidence interval [CI]: 0.91-1.71); median CSS was not reached, with the hazard ratio of death from melanoma 1.12 (95%CI: 0.81-1.75). The proportion of patients who were alive with melanoma recurrence or had died from melanoma was similar in both age groups. In multivariable analysis, higher pathological stage was the only independent risk factor for short RFS regardless of age group, sex, CCI, and tumor ulceration. Higher stage and male sex were associated with short CSS. Age ≥ 80 years, stage III disease, and CCI ≥ 2 were associated with short OS and female sex with long OS in multivariable analysis.

DISCUSSION

Pathological stage was the most influential factor determining RFS and CSS in older adults with resected stage I - III melanoma. Concerning OS, age ≥ 80 years, stage III disease, and multiple comorbidities had a significant negative impact.

Common variable immunodeficiency-an independent risk factor for atherosclerotic cardiovascular diseases

Atherosclerosis, a disease of chronic inflammation of the arterial wall, is the main cause of most cardiovascular diseases (CVDs). Common variable immunodeficiency (CVID), a group of diseases characterized by frequent infections due to defective antibody production and lack of human immunoglobulins, plays a role in immune activation and inflammation. Thus, it can be hypothesized that CVID increases the risk for atherosclerotic CVDs. On the other hand, it is also possible that CVID patients are protected from atherosclerotic CVDs based on their life-long immunoglobulin therapy. Here, we examined whether patients with CVID have an increased risk for atherosclerotic CVDs or whether they are protected from these diseases. Using an electronic patient database registry search of a population of 83 CVID patients and their age- and sex-matched, tenfold larger control population we demonstrate that CVID patients have a statistically significantly higher risk for coronary heart disease (OR 2.4, p = 0.015) and peripheral vascular disease (OR 12.5, p < 0.001). Regarding cerebrovascular disease, there was a trend towards CVID patients having more strokes or ischemic attacks, but the difference was not statistically significant (OR 2.0, p = 0.133). The combined OR for CVID patients for atherosclerotic CVDs was 2.6 (p = 0.001). CVID population had more hypertension, but smoking was more seldom. There were no statistically significant differences in the incidence of diabetes or levels of serum total, HDL or LDL cholesterol, or glycosylated hemoglobin A1c between CVID patients and their controls. CVID patients had infections more frequently and the OR for autoimmune diseases was 3.8 (p < 0.001). Finally, a multivariate logistic analysis showed that CVID is an independent risk factor for atherosclerotic CVDs (p = 0.002). The present study demonstrates for the first time that CVID is an independent risk factor for atherosclerotic CVDs. Further studies are required to fully understand the exact mechanisms behind this.

Artificial Intelligence based wrapper for high dimensional feature selection

BACKGROUND

Feature selection is important in high dimensional data analysis. The wrapper approach is one of the ways to perform feature selection, but it is computationally intensive as it builds and evaluates models of multiple subsets of features. The existing wrapper algorithm primarily focuses on shortening the path to find an optimal feature set. However, it underutilizes the capability of feature subset models, which impacts feature selection and its predictive performance.

METHOD AND RESULTS

This study proposes a novel Artificial Intelligence based Wrapper (AIWrap) algorithm that integrates Artificial Intelligence (AI) with the existing wrapper algorithm. The algorithm develops a Performance Prediction Model using AI which predicts the model performance of any feature set and allows the wrapper algorithm to evaluate the feature subset performance in a model without building the model. The algorithm can make the wrapper algorithm more relevant for high-dimensional data. We evaluate the performance of this algorithm using simulated studies and real research studies. AIWrap shows better or at par feature selection and model prediction performance than standard penalized feature selection algorithms and wrapper algorithms.

CONCLUSION

AIWrap approach provides an alternative algorithm to the existing algorithms for feature selection. The current study focuses on AIWrap application in continuous cross-sectional data. However, it could be applied to other datasets like longitudinal, categorical and time-to-event biological data.

The impact of smoking on recurrence and progression of non-muscle invasive bladder cancer: a systematic review and meta-analysis

OBJECTIVES

Although smoking is a well-recognized causative factor of urothelial bladder cancer and accounts for 50% of cases, less is known about the prognostic significance of smoking on non-muscle invasive bladder cancer (NMIBC) prognosis. This systematic review and meta-analysis aimed to evaluate the effect of smoking on the risk of NMIBC recurrence and progression.

MATERIALS AND METHODS

We systematically searched Medline, Web of Science and Scopus databases for original articles published before October 2021 regarding the effect of smoking on NMIBC recurrence and progression. Information about smoking status and the number of events or odds ratio or hazard ratio for event-free survival must have been reported to include the study in the analysis. Quality In Prognosis Studies tool was utilized for the risk of bias assessment.

RESULTS

We selected 64 eligible studies, including 28 617 patients with NMIBC with available data on smoking status. In a meta-analysis of 28 studies with 7885 patients, we found that smokers (current/former) were at higher risk for recurrence (OR = 1.68; 95% CI 1.34-2.09; P < 0.0001) compared to never smokers. Subgroup analysis of 2967 patients revealed that current smokers were at a 1.24 higher risk of recurrence (OR = 1.24; 95% CI 1.02-1.50; P = 0.03) compared to former smokers. A meta-analysis of the hazard ratio revealed that smokers are at higher risk of recurrence (HR = 1.31; 95%CI 1.15-1.48; P < 0.0001) and progression (HR = 1.18; 95%CI 1.08-1.29; P < 0.001) compared to never smokers. Detrimental prognostic effect of smoking on progression, but not for recurrence risk was also noted in the subgroup analysis of high-risk patients (HR = 1.30; 95%CI 1.09-1.55; P = 0.004) and BCG-treated ones (HR = 1.15; 95%CI 1.06-1.25; P < 0.001).

CONCLUSION

In conclusion, patients with non-muscle invasive bladder cancer and a history of smoking have a worse prognosis regarding recurrence-free and progression-free survival compared to non-smokers.

The clinical impact of continued smoking in patients with breast and other hormone-dependent cancer: A systematic literature review

We conducted a systematic review of studies that investigated whether quitting smoking at or around diagnosis improves survival of patients with hormone-dependent cancers (HDC). Nine studies published in 2013-2022 were included. Studies were very diverse in terms of design, definition of quitters and continued smokers, and prevalence of prognostic factors other than smoking cessation (e.g. patients' demographics, tumour characteristic, and treatments). For breast, ovarian, and endometrial cancer, all included studies found that quitters had better overall, disease specific, and disease-free survival than continued smokers. For prostate cancer, there was no evidence of an association of smoking cessation with improved survival. This literature review provided suggestive evidence that female smokers diagnosed with cancer of the breast, ovary, or endometrium may improve their chances of surviving by stopping smoking. Smoking cessation counselling should become part of standard oncological care for these patients and integrated into breast cancer screening programs.

Smoking is an Independent Marker of Poor Prognosis in Cutaneous Melanoma

Previous studies have suggested that persistent tobacco smoking impairs survival in cutaneous melanoma, but the effects of smoking and other prognostic factors have not been described in detail. This study examined the association of smoking (persistent, former, or never) with melanoma-specific (MSS) and overall survival (OS) in patients with cutaneous melanoma treated in Southwest Finland during 2005 to 2019. Clinical characteristics were obtained from electronic health records for 1,980 patients. Smoking status was available for 1,359 patients. Patients were restaged according to the 8th edition of the tumour-node-metastasis (TNM) classification. Smoking remained an independent prognostic factor for inferior melanoma-specific survival regardless of age, sex, stage, and comorbidities. The hazard ratio of death from melanoma was 1.81 (1.27-2.58, p = 0.001) in persistent and 1.75 (1.28-2.40, p = 0.001) in former smokers compared with never smokers. In 351 stage IV patients, smoking was associated with increased melanoma-specific and overall mortality: median MSS 10.4 (6.5-14.3), 14.6 (9.1-20.1), and 14.9 (11.4-18.4) months, p = 0.01 and median OS 10.4 (6.5-14.3), 13.9 (8.6-19.2), and 14.9 (11.7-18.1) months, p = 0.01 in persistent, former, and never smokers, respectively. In conclusion, since smoking represents an independent modifiable poor prognostic factor in patients with cutaneous melanoma, smoking habits should be proactively asked about by healthcare professionals, in order to support smoking cessation.

Clinical Characteristics and Disease Course of Fibrosing Interstitial Lung Disease Patients in a Real-World Setting

Background and Objectives: This single-center retrospective study was conducted to describe clinical characteristics and the disease course of patients with interstitial lung diseases (ILD). Materials and Methods: The study included adult patients with fibrosing ILD (IPF, non-IPF fibrosing ILD (F-ILD), and non-IPF progressive pulmonary fibrosis (PPF)) treated between 2014 and 2017. Patients were followed annually from the first visit until the end of the study period in 2019. Data were collected from the Turku University Hospital data lake and analyzed using descriptive statistics. Results: 591 patients formed the patient cohort: 110 had IPF, 194 F-ILD, 142 PPF, and the remaining 145 patients were uncertain, F-ILD-U, whose disease progression nature could not be confirmed by FVC measurements. There were more males in each patient group and median age of the groups was similar, although there were younger patients in the PPF, F-ILD, and F-ILD-U groups. PPF patients had more UIP pattern than F-ILD patients. Exposure-related ILDs were clearly the most found ILD diagnoses for both PPF and F-ILD, followed by unclassifiable IIP. Baseline FVC % predicted reduction in every group was moderate. Half of the patients in each group had comorbidities, and the most common were cardiovascular diseases, diabetes, sleep apnea, and chronic lower respiratory diseases; F-ILD-U patients had malignant diseases as well. IPF patients had less medications than the other groups. Glucocorticoids were the most used medication in all patient groups. More PPF and F-ILD patients remained in the follow-up than IPF and F-ILD-U patients. Similarly, mortality of F-ILD-U was the highest, followed by IPF. Evolvement of lung function, oxygen use, and number of acute hospitalizations were similar for IPF and PPF patients whereas the corresponding results were always better for F-ILD patients. Conclusions: The disease course of IPF and PPF was similar, and PPF patient amount exceeded the amount of IPF patients.

Controlled register‐based study of road traffic accidents in 12,651 Finnish cancer patients during 2013–2019

BACKGROUND

Little controlled evidence exists on road traffic accident (RTA) risk among patients diagnosed with cancer, while clinicians are often requested to comment their ability to drive. The aim of this study was to evaluate RTA risk in a population-based cohort of cancer patients living in Southwest Finland.

PATIENTS

All adult patients diagnosed with cancer in 2013-2019 were included. Acute appendectomy/cholecystectomy and actinic keratosis patients without cancer were selected from the same region as the control cohort. Participants were cross-referenced to a national driving licence database, yielding 12,651 cancer and 6334 control patients with a valid licence. Due to marked differences in their clinical presentation, the cancer cohort was divided into nine cancers of interest (breast, prostate, colorectal, lung, melanoma, head & neck, primary brain tumours, gynaecological and haematological malignancies). The nationwide law-regulated motor liability insurance registry was searched for all RTAs leading to injury with claims paid to not- or at-fault participants. At-fault drivers were verified based on sex and birth year.

RESULTS

During a median follow-up of 34 months, 167 persons were at-fault drivers in RTAs leading to injury. Among the nine cancers of interest, RTA risk did not differ from the control cohort. Among cancer patients, multivariable regression suggested male sex and opioid use, but not advanced cancer stage or given systemic therapy, as the most influential risk factors for RTA.

CONCLUSIONS

Cancer diagnosis itself was not associated with increased RTA risk, but other associated symptoms, medications, comorbidities or specific cancer subgroups may.

[Risk Factors and Pathogenic Mechanism for Secondary Primary Lung Cancer in Breast Cancer Patients: A Review]

Breast cancer and lung cancer are the two most common malignancies in China. With the continuous improvement of breast cancer diagnosis and treatment technology, the survival time of breast cancer patients has been prolonged, and the number of breast cancer patients with second primary lung cancer (SPLC) has increased. In addition, breast cancer is the most common primary cancer in female patients with SPLC, and SPLC is the main cause of death in this population. More and more physicians pay attention to this clinical phenomenon. This paper summarized the risk and risk factors of SPLC in breast cancer patients, and elaborated its pathogenesis, in order to provide a theoretical basis for the clinical management of breast cancer patients and achieve accurate early intervention as soon as possible.
.

Changes in Cigarette Smoking Behavior in Cancer Survivors During Diagnosis and Treatment

INTRODUCTION

Continued cigarette smoking is a critical determinant of outcome in oncology patients, but how the separate events of cancer diagnosis, cancer treatment, and completion of treatment affect smoking behavior in cancer survivors is unknown. This study described such changes, hypothesizing that they would be more pronounced for smoking-related cancers.

METHODS

The Cancer Patient Tobacco Use Questionnaire (CTUQ) was sent to every cancer patient ≥ 18 y old scheduled for outpatient visits at the Mayo Clinic Cancer Center with a current or former history of tobacco use.

RESULTS

From September 2019 to September 2020, 33,831 patients received the CTUQ, and 20,818 (62%) responded. Of the 3007 current smokers analyzed, 34% quit at diagnosis; those with smoking-related cancers were more likely to quit (40% vs. 29%, respectively, p < .001). Among those who did not quit at diagnosis, 31% quit after starting cancer treatment, and those with smoking-related cancers were more likely to quit (35% and 28%, respectively, p = .002) Among those who had quit before the completion of treatment, 13% resumed smoking after treatment ended. In multivariable analysis, patients with smoking-related cancers were more likely to report 30-d point prevalence abstinence from pre-diagnosis to the completion of treatment (adjusted OR 1.98 [95% CI 1.65, 2.36], p < .001).

CONCLUSIONS

Both cancer diagnosis and treatment prompt smoking abstinence, with most maintaining abstinence after treatment is completed. Those with smoking-related cancers are more likely to quit. These results emphasize the need to provide access to tobacco treatment services that can further support cancer patients who smoke.

IMPLICATIONS

Approximately one-third of cancer patients who smoke quit smoking at the time of diagnosis, an additional one-third of patients who had not yet quit did so at the initiation of treatment, and most sustained this abstinence after treatment completion. Patients with smoking-related cancers are more likely to quit. Thus, there are multiple opportunities within the cancer care continuum to intervene and support quit attempts or continued abstinence.

Prognostic Impact of Post-Diagnosis Smoking Cessation among Bladder Cancer Patients: A Systematic Literature Review and Meta-Analysis

We reviewed the studies examining whether quitting smoking at or around diagnosis favourably affects the prognosis of bladder cancer (BC) patients, who are often active smokers at diagnosis. We found only nine eligible articles published until 31 January 2022, which encompassed around 5500 BC in total, the majority of which were nonmuscle invasive BC (only one paper included muscle-invasive BC). We used random effects meta-analysis to obtain a summary hazard ratio (SHR) and 95% confidence intervals (CI). The median proportion of smokers who quit at or around diagnosis was 29.8% (range 8.4-43.1%). For the overall, BC-specific, and progression-free survival, the studies were limited in number (n = 3) and provided conflicting results. At the same time, quitters did not appear to have a lower risk of recurrence than continued smokers (SHR 0.99, 95% CI 0.61-1.61). In conclusion, while the evidence is currently not sufficient to draw firm conclusions (especially for patients with muscle-invasive BC), physicians should not refrain from educating smoking BC patients about the benefits of smoking cessation and provide the necessary support.

Quitting Smoking At or Around Diagnosis Improves the Overall Survival of Lung Cancer Patients: A Systematic Review and Meta-Analysis

INTRODUCTION

Lung cancer (LC) remains a disease with poor prognosis despite recent advances in treatments. Here, we aimed at summarizing the current scientific evidence on whether quitting smoking at or around diagnosis has a beneficial effect on the survival of LC patients.

METHODS

We searched MEDLINE and EMBASE for articles published until 31st October, 2021, that quantified the impact on LC patients' survival of quitting smoking at or around diagnosis or during treatment. Study-specific data were pooled into summary relative risk (SRR) and corresponding 95% confidence intervals (CI) using random effect meta-analysis models.

RESULTS

Twenty-one articles published between 1980 and 2021 were included, which encompassed a total of over 10,000 LC patients. There was substantial variability across studies in terms of design, patients' characteristics, treatments received, criteria used to define smoking status (quitters or continued), and duration of follow-up. Quitting smoking at or around diagnosis was significantly associated with improved overall survival (SRR 0.71, 95% CI 0.64-0.80), consistently among patients with non-small cell LC (SRR 0.77, 95% CI 0.66-0.90, n studies = 8), small cell LC (SRR 0.75, 95% CI 0.57-0.99, n studies = 4), or LC of both or unspecified histological type (SRR 0.81, 95% CI 0.68-0.96, n studies = 6).

CONCLUSIONS

Quitting smoking at or around diagnosis is associated with a beneficial effect on the survival of LC patients. Treating physicians should educate LC patients about the benefits of quitting smoking even after diagnosis and provide them with the necessary smoking cessation support.

Documentation of the patient's smoking status in common chronic diseases - analysis of medical narrative reports using the ULMFiT based text classification

INTRODUCTION

Smoking cessation is essential part of a successful treatment in many chronic diseases. Our aim was to analyse how actively clinicians discuss and document patients' smoking status into electronic health records (EHR) and deliver smoking cessation assistance.

METHODS

We analysed the results using a combination of rule and deep learning-based algorithms. Narrative reports of all adult patients, whose treatment started between years 2010 and 2016 for one of seven common chronic diseases, were followed for two years. Smoking related sentences were first extracted with a rule-based algorithm. Subsequently, pre-trained ULMFiT-based algorithm classified each patient's smoking status as a current smoker, ex-smoker, or never smoker. A rule-based algorithm was then again used to analyse the physician-patient discussions on smoking cessation among current smokers.

RESULTS

A total of 35,650 patients were studied. Of all patients, 60% were found to have a smoking status in EHR and the documentation improved over time. Smoking status was documented more actively among COPD (86%) and sleep apnoea (83%) patients compared to patients with asthma, type 1&2 diabetes, cerebral infarction and ischemic heart disease (range 44-61%). Of the current smokers (N=7,105), 49% had discussed smoking cessation with their physician. The performance of ULMFiT-based classifier was good with F-scores 79-92.

CONCLUSION

Ee found that smoking status was documented in 60% of patients with chronic disease and that the clinician had discussed smoking cessation in 49% of patients who were current smokers. ULMFiT-based classifier showed good/excellent performance and allowed us to efficiently study a large number of patients' medical narratives.

Smoking, DNA Methylation, and Breast Cancer: A Mendelian Randomization Study

Background

Smoking was strongly associated with breast cancer in previous studies. Whether smoking promotes breast cancer through DNA methylation remains unknown.

Methods

Two-sample Mendelian randomization (MR) analyses were conducted to assess the causal effect of smoking-related DNA methylation on breast cancer risk. We used 436 smoking-related CpG sites extracted from 846 middle-aged women in the ARIES project as exposure data. We collected summary data of breast cancer from one of the largest meta-analyses, including 69,501 cases for ER+ breast cancer and 21,468 cases for ER- breast cancer. A total of 485 single-nucleotide polymorphisms (SNPs) were selected as instrumental variables (IVs) for smoking-related DNA methylation. We further performed an MR Steiger test to estimate the likely direction of causal estimate between DNA methylation and breast cancer. We also conducted colocalization analysis to evaluate whether smoking-related CpG sites shared a common genetic causal SNP with breast cancer in a given region.

Results

We established four significant associations after multiple testing correction: the CpG sites of cg2583948 [OR = 0.94, 95% CI (0.91-0.97)], cg0760265 [OR = 1.07, 95% CI (1.03-1.11)], cg0420946 [OR = 0.95, 95% CI (0.93-0.98)], and cg2037583 [OR =1.09, 95% CI (1.04-1.15)] were associated with the risk of ER+ breast cancer. All the four smoking-related CpG sites had a larger variance than that in ER+ breast cancer (all p < 1.83 × 10^-11) in the MR Steiger test. Further colocalization analysis showed that there was strong evidence (based on PPH4 > 0.8) supporting a common genetic causal SNP between the CpG site of cg2583948 [with IMP3 expression (PPH4 = 0.958)] and ER+ breast cancer. There were no causal associations between smoking-related DNA methylation and ER- breast cancer.

Conclusions

These findings highlight potential targets for the prevention of ER+ breast cancer. Tissue-specific epigenetic data are required to confirm these results.