Validation of a nomogram for predicting risk of local recurrence for ductal carcinoma in situ

A Nomogram Model for Stratifying the Risk of Recurrence in Patients with Meningioma After Surgery

BACKGROUND

Here, we aimed to investigate the clinical parameters affecting the recurrence of meningiomas, and to construct a predictive nomogram model, so as to predict the recurrence-free survival (RFS) of meningiomas more accurately.

METHODS

The Clinical, imaging, and pathological data of 155 primary meningioma patients treated surgically from January 2014 to March 2021 were retrospectively analyzed. Independent prognostic factors affecting postoperative recurrence of meningioma were identified by univariate and multivariate Cox regression analyses. A predictive nomogram was established based on independent influence parameters. Subsequently, time-dependent receiver operating characteristic curve, calibration curve, and Kaplan-Meier method were utilized to evaluate the predictive ability of the model.

RESULTS

The multivariate Cox regression analysis showed that tumor size, Ki-67 index, and resection extent had independent prognostic significance, and these parameters were subsequently used to construct a predictive nomogram. Receiver operating characteristic curves indicated that the model was more accurate in predicting RFS than independent factors. Calibration curves suggested that the predicted RFS were similar to the actual observed RFS. In the Kaplan-Meier analysis, the RFS of high-risk cases was obviously shorter than that of low-risk cases.

CONCLUSIONS

The tumor size, Ki-67 index, and extent of resection were independent factors affecting the RFS of meningioma. The predictive nomogram based on these factors can be used as an effective method to stratify the recurrence risk of meningioma and provide a reference for patients to choose personalized treatment.

A nomogram for predicting brain metastases of EGFR-mutated lung adenocarcinoma patients and estimating the efficacy of therapeutic strategies

Background

To establish a nomogram for predicting the outcome of EGFR-mutated lung adenocarcinoma patients with brain metastases (BMs) and to estimate the efficacy of different therapeutic strategies.

Methods

The data of 129 cases with BM from the period between January 1st 2011 and December 31st 2014 were collected, and all of the cases were pathologically confirmed to be lung adenocarcinoma, stages I–IV and with 19 and/or 21 exon mutations of EGFR. Cox regression analysis and log-rank test were used for data analysis. The nomogram was used to establish the progression models.

Results

In the univariate analysis, the stage, ECOG score, interval between the diagnosis of lung cancer and BM, the number of brain metastatic lesions, and the diameter of the maximal brain metastatic lesion correlated well with overall survival (OS). In multivariate Cox proportional hazard analysis, the ECOG score, interval between the diagnosis of lung cancer and BM, and the number of brain metastatic lesions correlated well with the OS. Patients were divided into the poor prognostic group and the good prognostic group based on the nomogram prognostic model score. Subgroup analysis showed that in the poor prognostic group, the OS of patients who received radiotherapy was better than that of the patients who did not receive radiotherapy as the first-line treatment (30 vs. 19 months, P<0.05). The OS was 30 months in the TKI subgroup and 21 months in the no TKI subgroup, but no statistical difference was found (P>0.05). Patients in the good prognostic group who received radiotherapy had a better 3-y OS rate than the patients who received no radiotherapy as the first-line treatment (91.2% vs. 58.1%, P<0.05). The 3-y OS rate was 87.6% in the TKI subgroup and 67.8% in the no TKI group (P<0.05).

Conclusions

We established an effective nomogram model to predict the progression of EGFR-mutated lung adenocarcinoma patients with BM and the therapeutic effect of the individual treatments. Radiotherapy was beneficial for the patients of both the poor and good prognostic groups, but TKI may be better suited for treating the patients with good prognosis.

Risk Prediction for Local Breast Cancer Recurrence Among Women with DCIS Treated in a Community Practice: A Nested, Case-Control Study

BACKGROUND

Various patient, treatment, and pathologic factors have been associated with an increased risk of local recurrence (LR) following breast-conserving therapy (BCT) for ductal carcinoma in situ (DCIS). However, the strength and importance of individual factors has varied; whether combining factors improves prediction, particularly in community practice, is uncertain. In a large, population-based cohort of women with DCIS treated with BCT in three community-based practices, we assessed the validity of the Memorial Sloan-Kettering Cancer Center (MSKCC) DCIS nomogram, which combines clinical, pathologic, and treatment features to predict LR.

METHODS

We reviewed slides of patients with unilateral DCIS treated with BCT. Regression methods were used to estimate risks of LR. The MSKCC DCIS nomogram was applied to the study population to compare the nomogram-predicted and observed LR at 5 and 10 years.

RESULTS

The 495 patients in our study were grouped into quartiles and octiles to compare observed and nomogram-predicted LR. The 5-year absolute risk of recurrence for lowest and highest quartiles was 4.8 and 33.1 % (95 % CI 3.1-6.4 and 24.2-40.9, respectively; p < 0.0001). The overall correlation between 10-year nomogram-predicted recurrences and observed recurrences was 0.95. Compared with observed 10-year LR rates, the risk estimates provided by the nomogram showed good correlation, and reasonable discrimination with a c-statistic of 0.68.

CONCLUSIONS

The MSKCC DCIS nomogram provided good prediction of the 5- and 10-year LR when applied to a population of patients with DCIS treated with BCT in a community-based practice. This nomogram, therefore, is a useful treatment decision aid for patients with DCIS.