Construction, validation, and visualization of a web-based nomogram to predict survival in male breast cancer patients with second primary prostate cancer

Abstract

Background: The advancement of early detection and treatment has brought about a significant concern for male breast cancer (MBC) survivors-the emergence of a second primary malignancy (SPM) poses a grave threat to their lives. Among them, second primary prostate cancer (spPCa) holds particular significance. This study aimed to investigate the impact of spPCa on the prognosis of MBC patients.

Methods: We performed a retrospective analysis using information from the Surveillance, Epidemiology, and End Results (SEER) database to investigate individuals diagnosed with MBC who also experienced an SPM between 2000 and 2020. Propensity score matching (PSM) was employed to balance the baseline characteristics of individuals with spPCa and those with second primary non-prostate cancer (non-PCa). The impact of spPCa on participant survival was assessed using the Kaplan-Meier method. Furthermore, two nomograms were developed, based on univariate and multifactor Cox regression analyses, to predict overall survival (OS) and cancer-specific survival (CSS). The capacity of the nomograms was evaluated using the concordance index (C-index), calibration curve, receiver operating characteristic (ROC) analysis, and decision curve analysis (DCA). Additionally, a risk stratification system was devised, taking into account the cumulative score of each patient in the nomogram.

Results: This study enrolled a total of 885 MBC patients who experienced an SPM, of which 265 (29.9%) were diagnosed with spPCa. Through PSM, 257 pairs of eligible participants were selected. Survival analysis revealed that patients with prostate cancer (PCa) as an SPM have longer OS and CSS compared to those with other types of cancer as an SPM. The participants were randomly divided into a training set and a validation set in a ratio of 7:3. The Cox proportional hazards model was utilized to assess the risk factors associated with survival outcomes. Two nomograms were developed to forecast the 3-, 5-, 8-, and 10-year OS and CSS of male patients who had breast cancer and SPM. The two nomograms exhibited excellent performance in terms of the C-index, ROC curves, calibration plots, and DCA curves, demonstrating their exceptional clinical discriminative ability and predictive utility. In the risk stratification system predicated on the total score of the nomogram, patients deemed high-risk exhibited diminished OS and CSS. Additionally, we created user-friendly web applications to enhance the accessibility of the nomogram in clinical practices, which can be accessed at https://mbcpre.shinyapps.io/DynNomapp_OS/ for OS and https://mbcpre.shinyapps.io/DynNomapp_CSS/ for CSS.

Conclusions: MBC patients with spPCa exhibit a more favorable prognosis than those with other SPMs. The two nomograms we constructed could accurately forecast the OS and CSS for MBC patients with spPCa. Patients whose nomograms are stratified as high-risk should gain additional attention. Our nomograms may aid clinicians in personalizing treatment strategies and supporting clinical decisions.

Keywords: Male breast cancer (MBC); Surveillance, Epidemiology, and End Results (SEER); nomogram; second primary prostate cancer (spPCa); survival probability.

Figure 1

Flowchart of data screening and research process. SEER, Surveillance, Epidemiology, and End Results; MBC, male breast cancer; PSM, propensity score matching; OS, overall survival; CSS, cancer-specific survival; ROC, receiver operating characteristic; DCA, decision curve analysis; K-M, Kaplan-Meier.

Figure 2

Kaplan-Meier curves for OS between PCa and non-PCa patients before (A) and after (B) PSM. PCa, prostate cancer; OS, overall survival; PSM, propensity score matching.

Figure 3

Kaplan-Meier curves for CSS between PCa and non-PCa patients before (A) and after (B) PSM. PCa, prostate cancer; CSS, cancer-specific survival; PSM, propensity score matching.

Figure 4

The nomograms for predicting 3-, 5-, 8-, and 10- year OS (A) and CSS (B) in MBC patients with SPM. PC, prostate cancer; HER2, human epidermal growth factor receptor 2; PR, progesterone receptor; OS, overall survival; CSS, cancer-specific survival; MBC, male breast cancer; SPM, second primary malignancy; SDWU, separated, divorced, widowed, or unmarried.

Figure 5

Calibration curves of nomograms for predicting the 3-, 5-, 8-, and 10-year OS and CSS in MBC patients with SPM. OS in the training set (A) and the validation set (B); CSS in the training set (C) and the validation set (D). MBC, male breast cancer; SPM, second primary malignancy; OS, overall survival; CSS, cancer-specific survival.

Figure 6

ROC of nomograms for predicting the 3-, 5-, 8-, and 10-year OS and CSS in MBC patients with SPM. OS in the training set (A) and the validation set (B); CSS in the training set (C) and the validation set (D). MBC, male breast cancer; SPM, second primary malignancy; OS, overall survival; CSS, cancer-specific survival; ROC, receiver operating characteristic; AUC, area under the curve; CI, confidence interval.

Figure 7

DCA of nomograms for predicting the 3-, 5-, 8-, and 10-year OS and CSS in MBC patients with SPM. OS in the training set (A) and the validation set (B); CSS in the training set (C) and the validation set (D). DCA, decision curve analysis; MBC, male breast cancer; SPM, second primary malignancy; OS, overall survival; CSS, cancer-specific survival; AJCC, American Joint Committee on Cancer.

Figure 8

The Kaplan-Meier curves among different risk groups in nomogram predicting model of OS (A) and CSS (B). OS, overall survival; CSS, cancer-specific survival.