Understanding and identifying immortal-time bias in surgical health services research: An example using surgical resection of stage IV breast cancer

Surgery on Metastatic Foci is a Better Strategy for Stage IV Breast Cancer Patients with only Nonregional Lymph Node Metastasis

INTRODUCTION

Our study aims to explore whether breast cancer patients with non-regional lymph node (NRLN) metastasis benefit from surgery on distant nodes, and to determine the influencing factors affecting the prognosis of this type of patient.

METHODS

Information of invasive ductal carcinoma (IDC) patients from 2004 to 2016 was extracted from the Surveillance, Epidemiology, and End Results (SEER) database and analyzed by statistical methods, including multivariate Cox regression model, chi-squared test, propensity score matching (PSM), Kaplan-Meier plot, and log-rank test.

RESULTS

A total of 4236 M1 patients met the designated criteria. Among 847 patients with only NRLN metastasis who have detailed information, only 114 patients received surgery on metastatic distant lymph nodes. The Kaplan-Meier plot for overall survival (OS) showed that the prognosis of NRLN metastatic patients was superior to visceral metastasis (P < 0.0001) but similar to supraclavicular metastasis (P = 0.33). In addition, NRLN metastatic patients who underwent surgery on the NRLNs were found to have superior prognoses in terms of both OS (P = 0.041) and cancer-specific survival (P = 0.034) compared with those who did not undergo NRLN surgery. We have also demonstrated that NRLN metastatic patients who have received radiotherapy plus chemotherapy for primary tumors gain superior survival compared with those who only received chemotherapy apart from NRLN surgery.

CONCLUSION

Surgery on NRLN and radiotherapy for the primary tumor improved the prognosis of NRLN metastatic patients. Thus, the classification of NRLN, especially contralateral axillary lymph node metastasis (CAM), into the M1 breast cancer stage should be reconsidered. Different locoregional treatment strategies for metastatic foci should be recommended for patients with only NRLN and patients with visceral metastasis.

The Timing, the Treatment, the Question: Comparison of Epidemiologic Approaches to Minimize Immortal Time Bias in Real-World Data Using a Surgical Oncology Example

BACKGROUND

Studies evaluating the effects of cancer treatments are prone to immortal time bias that, if unaddressed, can lead to treatments appearing more beneficial than they are.

METHODS

To demonstrate the impact of immortal time bias, we compared results across several analytic approaches (dichotomous exposure, dichotomous exposure excluding immortal time, time-varying exposure, landmark analysis, clone-censor-weight method), using surgical resection among women with metastatic breast cancer as an example. All adult women diagnosed with incident metastatic breast cancer from 2013-2016 in the National Cancer Database were included. To quantify immortal time bias, we also conducted a simulation study where the "true" relationship between surgical resection and mortality was known.

RESULTS

24,329 women (median age 61, IQR 51-71) were included, and 24% underwent surgical resection. The largest association between resection and mortality was observed when using a dichotomized exposure [HR, 0.54; 95% confidence interval (CI), 0.51-0.57], followed by dichotomous with exclusion of immortal time (HR, 0.62; 95% CI, 0.59-0.65). Results from the time-varying exposure, landmark, and clone-censor-weight method analyses were closer to the null (HR, 0.67-0.84). Results from the plasmode simulation found that the time-varying exposure, landmark, and clone-censor-weight method models all produced unbiased HRs (bias -0.003 to 0.016). Both standard dichotomous exposure (HR, 0.84; bias, -0.177) and dichotomous with exclusion of immortal time (HR, 0.93; bias, -0.074) produced meaningfully biased estimates.

CONCLUSIONS

Researchers should use time-varying exposures with a treatment assessment window or the clone-censor-weight method when immortal time is present.

IMPACT

Using methods that appropriately account for immortal time will improve evidence and decision-making from research using real-world data.

Trimodality Therapy vs Definitive Chemoradiation in Older Adults With Locally Advanced Esophageal Cancer

BACKGROUND

The comparative effectiveness of trimodality therapy vs definitive chemoradiation for treating locally advanced esophageal cancer in older adults is uncertain. Existing trials lack generalizability to older adults, a population with heightened frailty. We sought to emulate a hypothetical trial comparing these treatments using real-world data.

METHODS

A cohort of adults aged 66-79 years diagnosed with locally advanced esophageal cancer between 2004 and 2017 was identified in the Surveillance Epidemiology and End Results-Medicare database. The clone-censor-weight method was leveraged to eliminate time-related biases when comparing outcomes between treatments. Outcomes included overall mortality, esophageal cancer-specific mortality, functional adverse events, and healthy days at home.

RESULTS

A total of 1240 individuals with adenocarcinomas and 661 with squamous cell carcinomas were identified. For adenocarcinomas, the standardized 5-year risk of mortality was 73.4% for trimodality therapy and 83.8% for definitive chemoradiation (relative risk [RR] = 0.88, 95% confidence interval [CI] = 0.82 to 0.95). Trimodality therapy was associated with mortality risk reduction for squamous cell carcinomas (RR = 0.87, 95% CI = 0.70 to 1.01). The 1-year incidence of functional adverse events was higher in the trimodality group (adenocarcinomas RR = 1.40, 95% CI = 1.22 to 1.65; squamous cell carcinomas RR = 1.21, 95% CI = 1.00 to 1.49). Over 5 years, trimodality therapy was associated with 160 (95% CI = 67 to 229) and 177 (95% CI = 51 to 313) additional home days in individuals with adenocarcinomas and squamous cell carcinomas, respectively.

CONCLUSIONS

Compared with definitive chemoradiation, trimodality therapy was associated with reduced mortality but increased risk of function-related adverse events. Discussing these tradeoffs may help optimize care plans.

Primary-Site Local Therapy for Patients with De Novo Metastatic Breast Cancer: An Educational Review

BACKGROUND

Until 2001, the paradigm guiding the management of women with de novo metastatic breast cancer (dnMBC) stipulated that primary-site locoregional therapy (PSLT) did not alter the course of metastatic disease and was necessary only for palliation of symptoms. Since 2002, retrospective data have begun questioning this paradigm. However, selection biases driving an observed survival advantage associated with PSLT in dnMBC were quickly recognized and led to several randomized clinical trials (RCTs) addressing this question.

METHODS AND RESULTS

Four published RCTs have since tested the value of PSLT added to systemic therapy (ST) or not, with overall survival (OS) as the primary end point. The results of three published trials show no OS benefit for the addition of PSLT: Indian Tata Memorial, U.S./Canada E2108, and Austrian POSYTIVE (although POSYTIVE did not reach full accrual). The fourth RCT (Turkey, MF07-01) shows an OS benefit for PSLT at 5 years (42 % vs 24 % in the ST arm; hazard ratio [HR], 0.66; 95 % confidence interval [CI], 0.49-0.88). However, the 5-year survival in the PSLT arm of MF07-01 is similar to that in both arms of E2108, suggesting that the worse survival in the ST arm of MF07-01 is a result of biologically worse disease (from imbalanced randomization). Locoregional control was improved by PSLT in all trials, but without improvement in quality of life.

CONCLUSIONS

The current evidence fails to refute the 20th century paradigm guiding management of de novo metastatic breast cancer. Discussion continues regarding the survival value of PSLT for patients with bone-only disease or oligometastases, but unbiased evidence is lacking.