Predicting Lapatinib Dose Regimen Using Machine Learning and Deep Learning Techniques Based on a Real-World Study

Lapatinib is used for the treatment of metastatic HER2(+) breast cancer. We aim to establish a prediction model for lapatinib dose using machine learning and deep learning techniques based on a real-world study. There were 149 breast cancer patients enrolled from July 2016 to June 2017 at Fudan University Shanghai Cancer Center. The sequential forward selection algorithm based on random forest was applied for variable selection. Twelve machine learning and deep learning algorithms were compared in terms of their predictive abilities (logistic regression, SVM, random forest, Adaboost, XGBoost, GBDT, LightGBM, CatBoost, TabNet, ANN, Super TML, and Wide&Deep). As a result, TabNet was chosen to construct the prediction model with the best performance (accuracy = 0.82 and AUC = 0.83). Afterward, four variables that strongly correlated with lapatinib dose were ranked via importance score as follows: treatment protocols, weight, number of chemotherapy treatments, and number of metastases. Finally, the confusion matrix was used to validate the model for a dose regimen of 1,250 mg lapatinib (precision = 81% and recall = 95%), and for a dose regimen of 1,000 mg lapatinib (precision = 87% and recall = 64%). To conclude, we established a deep learning model to predict lapatinib dose based on important influencing variables selected from real-world evidence, to achieve an optimal individualized dose regimen with good predictive performance.

‘Breast Cancer Resistance Likelihood and Personalized Treatment Through Integrated Multiomics’

In recent times, enormous progress has been made in improving the diagnosis and therapeutic strategies for breast carcinoma, yet it remains the most prevalent cancer and second highest contributor to cancer-related deaths in women. Breast cancer (BC) affects one in eight females globally. In 2018 alone, 1.4 million cases were identified worldwide in postmenopausal women and 645,000 cases in premenopausal females, and this burden is constantly increasing. This shows that still a lot of efforts are required to discover therapeutic remedies for this disease. One of the major clinical complications associated with the treatment of breast carcinoma is the development of therapeutic resistance. Multidrug resistance (MDR) and consequent relapse on therapy are prevalent issues related to breast carcinoma; it is due to our incomplete understanding of the molecular mechanisms of breast carcinoma disease. Therefore, elucidating the molecular mechanisms involved in drug resistance is critical. For management of breast carcinoma, the treatment decision not only depends on the assessment of prognosis factors but also on the evaluation of pathological and clinical factors. Integrated data assessments of these multiple factors of breast carcinoma through multiomics can provide significant insight and hope for making therapeutic decisions. This omics approach is particularly helpful since it identifies the biomarkers of disease progression and treatment progress by collective characterization and quantification of pools of biological molecules within and among the cancerous cells. The scrupulous understanding of cancer and its treatment at the molecular level led to the concept of a personalized approach, which is one of the most significant advancements in modern oncology. Likewise, there are certain genetic and non-genetic tests available for BC which can help in personalized therapy. Genetically inherited risks can be screened for personal predisposition to BC, and genetic changes or variations (mutations) can also be identified to decide on the best treatment. Ultimately, further understanding of BC at the molecular level (multiomics) will define more precise choices in personalized medicine. In this review, we have summarized therapeutic resistance associated with BC and the techniques used for its management.

Overnight fasting before lapatinib administration to breast cancer patients leads to reduced toxicity compared with nighttime dosing: a retrospective cohort study from a randomized clinical trial

Background

The bioavailability of lapatinib is affected by food, even following the 1 hour fast recommended by the package insert. We hypothesized that overnight fasting would minimize food‐drug interactions. Here, we investigated if lapatinib administration timing is associated with its tolerability, efficacy, and pharmacokinetics.

Methods

This is a retrospective cohort study utilizing the medical records of patients enrolled in the JBCRG‐16/Neo‐LaTH randomized phase 2 trial for breast cancer patients treated with lapatinib. Lapatinib administration timing was divided into three groups: before breakfast (BB), between meals (BM), and at bedtime (AB). Side effects (SE), treatment discontinuation rate (TDR), relative dose intensity (RDI), pathological complete response (pCR) rate, and lapatinib serum trough concentration were compared between groups.

Results

About 140 patients were included in this study: BB 15, BM 51, and AB 74. A reduced risk of diarrhea {adjusted hazard ratio (HR), 0.51, 95% confidence interval (CI), 0.27‐0.89, p = 0.018}, and rash {adjusted HR, 0.37; 95% CI, 0.17‐0.70, p = 0.002} was seen in BB versus AB. Fewer patients with low RDI (< 0.85/<0.6) were in the BB group (BB 13% / 0%, BM 22% / 3.9%, AB 24% / 14%, p = 0.70 / 0.11). pCR was not diminished (p = 0.75). BB group had the lowest serum lapatinib concentration and variability (mean ±SD were 0.35 ± 0.15, 0.65 ± 0.32, 0.96 ± 0.43 µg/ml).

Conclusions

Compared to bedtime administration, lapatinib administration after overnight fasting reduces its toxicity without diminishing its therapeutic efficacy.

Dual Blockade of HER-2 Provides a Greater Magnitude of Benefit in Patients With Hormone-Negative Versus Hormone-Positive Breast Cancer

The dual small molecule tyrosine kinase inhibitor lapatinib blocks both human epidermal growth factor receptor (HER-1) and human epidermal growth factor receptor 2 (HER-2) tyrosine kinase activity by binding reversibly to the ATP-binding site of the receptor's intracellular domain. Lapatinib, in combination with capecitabine, has been approved in 2007 for the treatment of patients with advanced HER-2⁺ breast cancer upon progressive disease following standard chemotherapy. Approval was also extended to the treatment of postmenopausal women with advanced hormone receptor (HR)-positive and HER-2-positive breast cancer in 2010. More recently, clinical trials that have investigated the efficacy of dual HER-2 blockade in both the metastatic and neoadjuvant breast cancer settings. For example, in 2013 the European Medicines Agency approved the combination of lapatinib and trastuzumab in HER-2⁺/HR^- patients. We review the efficacy results from dual HER-2 blockade studies and present new post hoc analysis efficacy data according to HR status. We show that dual blockade of HER-2 appears to provide a greater magnitude of benefit in the HR^- versus the HR⁺ subgroup of patients. Finally, we examine the potential of molecularly subtyping HER-2⁺ tumors using the PAM50 test as a predictor of response to treatment with the combination of trastuzumab and lapatinib.

Lapatinib-Related Rash and Breast Cancer Outcome in the ALTTO Phase III Randomized Trial

BACKGROUND

Previously we have shown that early development of rash is associated with a higher chance of achieving pathological complete response to neoadjuvant lapatinib. In the current analysis, we investigate its impact on survival in the ALTTO phase III adjuvant trial.

METHODS

In ALTTO, patients with human epidermal growth factor receptor 2 (HER2)-positive early breast cancer were randomly assigned to adjuvant trastuzumab, lapatinib, their sequence, or their combination for a total duration of one year. We evaluated whether the development of early lapatinib-related rash (ie, within 6 weeks) is associated with disease-free (DFS) and overall survival (OS). Landmark analysis at eight weeks and time-dependent analysis were tested in a multivariable model stratifying on trial's stratification factors. All statistical tests were two-sided.

RESULTS

Out of 6098 lapatinib-treated patients, 3973(65.2%) were included in the landmark analysis, of whom 1389 (35.0%) had developed early rash. After median follow-up of 4.5 years, the development of early rash was associated with a trend of improved DFS (multivariable: hazard ratio [HR] = 0.87, 95% confidence interval [CI] = 0.73 to 1.03,P= .10) and statistically significantly improved OS (multivariable: HR = 0.63, 95% CI = 0.48 to 0.82,P< .001) compared with subjects without early rash. Compared with patients randomly assigned to trastuzumab (n = 2051), patients who were randomly assigned to trastuzumab/lapatinib combination and developed early rash (n = 692) had superior DFS (multivariable: HR = 0.72, 95% CI = 0.55 to 0.92,P= .01) and OS (multivariable: HR = 0.59, 95% CI = 0.39 to 0.90,P= .01). Time-dependent analysis suggests that the occurrence of rash is predictive of lapatinib benefit, both when given in combination or sequential to trastuzumab.

CONCLUSIONS

Our results indicate that early development of rash identifies patients who derive superior benefit from lapatinib-based therapy.

Dual Blockade with AFatinib and Trastuzumab as NEoadjuvant Treatment for Patients with Locally Advanced or Operable Breast Cancer Receiving Taxane-Anthracycline Containing Chemotherapy-DAFNE (GBG-70)

PURPOSE

Dual anti-HER2 blockade with trastuzumab/pertuzumab or trastuzumab/lapatinib in combination with anthracycline/taxane-based chemotherapy can reach pathologic complete response (pCR) rates of up to 60% in HER2-positive breast cancer. The DAFNE (Dual blockade with AFatinib and trastuzumab as NEoadjuvant treatment) phase II study (NCT015591477) investigated a dual blockade with the irreversible pan-HER inhibitor afatinib and trastuzumab in this setting.

EXPERIMENTAL DESIGN

Participants with untreated, centrally HER2-positive breast cancer were treated for 6 weeks with afatinib (20 mg/d) and trastuzumab [(8) 6 mg/kg/3 weeks] alone; followed by 12-week treatment with paclitaxel (80 mg/m(2)/1 week), trastuzumab, and afatinib; followed by 12 weeks with epirubicin (90 mg/m(2)/3 weeks), cyclophosphamide (600 mg/m(2)/3 weeks), and trastuzumab before surgery. Primary objective was pCR rate, defined as ypT0/is ypN0. We expected a pCR rate of 70%; 65 patients were needed to exclude a rate of ≤55%.

RESULTS

pCR rate was 49.2% [90% confidence interval (CI), 38.5-60.1] in 65 treated patients. Patients with hormone receptor-negative (N = 19) or hormone receptor-positive (N = 46) tumors showed pCR rates of 63.2% and 43.5%, respectively (P = 0.153). Patients with (N = 9) or without (N = 56) lymphocyte predominant breast cancer (LPBC) showed pCR rates of 100% and 41.1%, respectively (P < 0.001). PCR rate was not different in patients with or without PIK3CA tumor mutations (P = 0.363). Clinical responses were seen in 96.3% of 54 evaluable patients, and breast conserving surgery was possible in 59.4% of 62 assessable patients. Most frequent nonhematologic grade 3-4 toxicities were diarrhea (7.7%), increased creatinine (4.6%), and infection (4.6%). One patient developed symptomatic congestive heart failure.

CONCLUSIONS

Neoadjuvant treatment with afatinib, trastuzumab, and chemotherapy showed acceptable tolerability, and a pCR rate comparable with that of other anti-HER2 doublets but below challenging expectations.