Identification of hand-foot syndrome from cancer patients' blog posts: BERT-based deep-learning approach to detect potential adverse drug reaction symptoms

Early detection and management of adverse drug reactions (ADRs) is crucial for improving patients' quality of life. Hand-foot syndrome (HFS) is one of the most problematic ADRs for cancer patients. Recently, an increasing number of patients post their daily experiences to internet community, for example in blogs, where potential ADR signals not captured through routine clinic visits can be described. Therefore, this study aimed to identify patients with potential ADRs, focusing on HFS, from internet blogs by using natural language processing (NLP) deep-learning methods. From 10,646 blog posts, written in Japanese by cancer patients, 149 HFS-positive sentences were extracted after pre-processing, annotation and scrutiny by a certified oncology pharmacist. The HFS-positive sentences described not only HFS typical expressions like "pain" or "spoon nail", but also patient-derived unique expressions like onomatopoeic ones. The dataset was divided at a 4 to 1 ratio and used to train and evaluate three NLP deep-learning models: long short-term memory (LSTM), bidirectional LSTM and bidirectional encoder representations from transformers (BERT). The BERT model gave the best performance with precision 0.63, recall 0.82 and f1 score 0.71 in the HFS user identification task. Our results demonstrate that this NLP deep-learning model can successfully identify patients with potential HFS from blog posts, where patients' real wordings on symptoms or impacts on their daily lives are described. Thus, it should be feasible to utilize patient-generated text data to improve ADR management for individual patients.

Symptom management: the utility of regional cooling for hand-foot syndrome induced by pegylated liposomal doxorubicin in ovarian cancer

PURPOSE

Hand-foot syndrome (HFS) is a major side effect of pegylated liposomal doxorubicin (PLD). Regional cooling during PLD infusion was shown to improve severe HFS. We investigated the utility of frozen gloves and socks (FGS) as a simpler cooling method.

METHODS

To evaluate the utility and safety of regional cooling with FGS for PLD-induced HFS, we retrospectively analyzed patients with advanced ovarian cancer who used FGS during PLD-containing regimens.

RESULTS

Ninety-six patients were analyzed. The incidence of HFS was 51% (≥ grade 2, 32%) in the PLD group and 38% (≥ grade 2, 6%) in the PLD + CBDCA group. The respective percentages of patients who underwent PLD dose modification/discontinuation were 41%/75% in the PLD group and 9%/30% in the PLD + CBDCA group. The reasons for discontinuation of PLD and PLD + CBDCA therapy were progressive disease, HFS, allergy, oral mucositis, and others. HFS was the only reason for PLD dose modification in both the PLD and PLD + CBDCA groups. The completion rate of FGS was 96%, with discontinuation in three cases due to pain from cooling.

CONCLUSIONS

Our study indicates that FGS is a safe, simple method with good tolerability. A prospective study is needed for further assessment.

Hand-foot skin reaction is a beneficial indicator of sorafenib therapy for patients with hepatocellular carcinoma: a systemic review and meta-analysis

BACKGROUND

Sorafenib remains the only standard first-line drug for advanced hepatocellular carcinoma (HCC). Hand-foot skin reaction (HFSR) is a very common side-effect in patients treated with sorafenib, and also affects the treatment schedule and quality of life. However, the association of HFSR and response of HCC to sorafenib remain unclear.

METHODS

Databases including PubMed, EMBASE, Web of Science and Cochrane Central Register of Controlled Trials were searched up to May 7th, 2017. Review Manager 5.3 software was adopted for performing meta-analyses, Newcastle-Ottawa Scale for assessing the bias of cohort studies, and GRADEprofler software for further assessing outcomes obtained from meta-analyses.

RESULTS

1478 articles were reviewed, and 12 cohort studies with 1017 participants were included in the analyses. The pooled hazard ratio (HR) of overall survival is 0.45 (95% confidence interval (CI) 0.36, 0.55; P < 0.00001; I2 = 35%). The pooled HR of time to progression is 0.41 (95% CI 0.28, 0.60; P < 0.00001; I2 = 0%). Patients suffering HFSR had significantly better outcomes from sorafenib therapy than those without HFSR.

CONCLUSIONS

The results indicate that HFSR is a beneficial indicator for HCC patients receiving sorafenib therapy. However, molecular mechanisms accounting for sorafenib-induced HFSR in HCC patients remain.

Development and Validation of the Korean Version of Hand-Foot Skin Reaction and Quality of Life Questionnaire (HF-QoL-K)

Given the growing number of cancer patients and the resulting increase in the administration of chemotherapeutic agents, convenient and effective methods for measuring the symptoms and quality of life associated with the hand-foot syndrome (HFS) are needed. Therefore, the aim of this study was to develop and validate the Korean version of the hand-foot skin reaction and quality of life questionnaire (HF-QoL-K), comprising a 20-item symptom domain and an 18-item daily activity domain. After we developed the HF-QoL-K, 209 Korean patients with gynecologic cancer who were undergoing chemotherapeutic agents relating the HFS were asked to fill in the questionnaire. The content validity, internal consistency reliability, and test-retest reliability were evaluated. The internal validity index, Cronbach's alpha coefficient, and intra-class correlation coefficient of the HF-QoL-K were 0.90, 0.958, and 0.825 (95% confidence interval [CI], 0.774-0.865), respectively. The scatter plot (Pearson correlation coefficient, 0.826) and the Bland-Altman plot for test-retest reliability were also acceptable. The HF-QoL-K instrument is a valid and reliable questionnaire for the measurement of the symptoms and quality of life in Korean cancer patients suffering HFS.

The Effect of Pyridoxine for Prevention of Hand-Foot Syndrome in Colorectal Cancer Patients with Adjuvant Chemotherapy Using Capecitabine: A Randomized Study

BACKGROUND/Aims: To determine the effect of the pyridoxine for prevention of hand-foot syndrome in colorectal cancer patients with adjuvant chemotherapy using capecitabine.

METHODOLOGY

Colorectal cancer patients scheduled for capecitabine chemotherapy as adjuvant setting were randomly assigned to with or without concurrent oral pyridoxine (60 mg/d) groups. Patients were monitored whether being a development of National Cancer Institute Common Toxicity Criteria grade 2 or worse HFS until chemotherapy ended.

RESULTS

Sixty patients were enrolled in this study. Relative dose intensity was 89.5% in total. The median number of chemotherapy cycles to grade 2 or worse HFS was four in both groups. Grade 2 or worse HES developed in 18 (60.0%) of 30 control patients and in 18 (60.0%) of 30 pyridoxine patients. The cumulative dose of capecitabine to grade 2 or worse HFS was not different between the two groups.

CONCLUSIONS

Pyridoxine is not effective in prevention of capecitabine-associated HFS.

[Assessment of hand-foot syndrome in cancer outpatients undergoing chemotherapy]

Capecitabine, an oral prodrug of 5 -fluorouracil, is a promising treatment for colorectal, breast, and gastric cancers, but often causes hand-foot syndrome(HFS), which is the most common dose-limiting toxicity. The aim of this study was to evaluate of the efficacy of the pharmacist in providing support at ambulatory therapy centers, especially for HFS. The HFS is a higher-incidence adverse event that may develop during chemotherapy with capecitabine. Once developed, the symptoms significantly impair quality of life(QOL), leading to a reduction in the dosage or discontinuation of the treatment. Patient symptoms may therefore increase in severity. This study was performed to analyze the treatment adherence and adverse events resulting from capecitabine therapy provided by pharmacists to cancer outpatients. All patients were prescribed vitamin B6(pyridoxine), which can help to reduce or prevent HFS. A lesser or milder extent of HFS was detected in patients who had used a moisturizer at the same time as the introduction of capecitabine therapy. Adherence to this approach will benefit the patients' selfcare in maintaining moisture retention, which is an important countermeasure for HFS. Additionally, early introduction of effective countermeasures for skin care, dose reduction, and rest periods is important for HFS management; in addition, team care support is dispensable. Our support system may be useful for management strategies for HFS. We suggest that improved quality of lif e is needed in cancer outpatients being treated with chemotherapy.

Development and validation of a prediction index for hand-foot skin reaction in cancer patients receiving sorafenib

BACKGROUND

This study describes a repeated measures prediction index to identify patients at high risk of ≥grade 2 hand-foot skin reaction (HFSR) before each week of sorafenib therapy.

METHODS

Data from 451 patients who received a sorafenib (400 mg bid) as part of a clinical trial were reviewed (Escudier B, Eisen T, Stadler WM et al. Sorafenib in advanced clear-cell renal-cell carcinoma. N Engl J Med 2007; 356: 125-134). Generalized estimating equations were used to develop the final risk model. A risk-scoring algorithm (range 0-58) was then derived from the final model coefficients. External validation was then carried out on a new sample of 1145 patients who received sorafenib under an expanded access program.

RESULTS

Pretreatment white blood cell count, female gender, good performance status, presence of lung and liver metastases and number of affected organs were predictors for ≥grade 2 HFSR. A nonlinear association between HFSR risk and treatment duration was also identified where risk was maximized at week 5 followed by a gradual decline. Before each week of therapy, patients with risk scores>40 would be considered at high risk for developing ≥grade 2 HFSR.

CONCLUSIONS

The application and planned continued refinement of this prediction tool will be an important source of patient-specific risk information for the development of moderate to severe HFSR.

Capecitabine-induced systemic lupus erythematosus and palmoplantar erythrodysesthesia

Capecitabine is emerging as an important drug in the treatment of metastatic breast and colorectal cancers. Marketed as Xeloda ®, this prodrug is taken orally and readily absorbed. It is novel in its increased convenience for patients, similar efficacy to the intravenous form of its active metabolite and its increased tolerability.1 We present a woman with metastatic breast cancer who presented with cutaneous abnormalities two months after starting treatment with capecitabine. Various dermatologic side effects have been attributed to capecitabine, often requiring cessation of the offending drug. We describe an unreported dermatological side effect of capecitabine therapy, systemic lupus erythematosus concurrent with palmoplantar erythrodysesthesia. As the use of this chemotherapeutic agent becomes more prevalent, it is important to recognize the range of its cutaneous side effects.

Capecitabine versus 5-fluorouracil/folinic acid as adjuvant therapy for stage III colon cancer: final results from the X-ACT trial with analysis by age and preliminary evidence of a pharmacodynamic marker of efficacy

BACKGROUND

This multicenter randomized trial compared oral capecitabine with bolus i.v. 5-fluorouracil (5-FU)/folinic acid (FA) as adjuvant therapy for stage III colon cancer.

PATIENTS AND METHODS

Patients were assigned to 24 weeks of capecitabine 1250 mg/m(2) twice daily on days 1-14 every 3 weeks or 5-FU/FA (Mayo Clinic regimen). The primary end point was disease-free survival (DFS).

RESULTS

The intent-to-treat population received capecitabine (n = 1004) or 5-FU/FA (n = 983). With a median follow-up of 6.9 years, capecitabine was at least equivalent to 5-FU/FA in terms of DFS [hazard ratio (HR) = 0.88; 95% confidence interval (CI) 0.77-1.01] and overall survival (OS) (HR = 0.86; 95% CI 0.74-1.01); the 95% CI upper limits were significantly less than the predefined noninferiority margins of 1.20 (P < 0.0001) and 1.14 (P < 0.001), respectively. This pattern was maintained in all subgroups, including patients aged ≥ 70 years. Preplanned multivariate analyses showed that capecitabine had statistically significant beneficial effects on DFS (P = 0.021) and OS (P = 0.020) versus 5-FU/FA. A post hoc analysis suggested that the occurrence of hand-foot syndrome may be associated with better outcomes in capecitabine recipients.

CONCLUSION

Oral capecitabine is an effective alternative to bolus 5-FU/FA as adjuvant treatment of patients with stage III colon cancer with efficacy benefits maintained at 5 years and in older patients.