The impact of delayed surgical treatment of suspected stage I ovarian cancer: lessons for a pandemic

Alice Darling,Benjamin Albright,Nicholas Mayne,Chi-Fu Jeffrey Yang,Andrew Berchuck,Haley Moss
DOI: https://doi.org/10.1016/s0090-8258(21)01213-0
IF: 5.304
2021-08-01
Gynecologic Oncology
Abstract:Objectives: During the COVID-19 pandemic, the Society of Gynecologic Oncology and other medical societies provided guidelines about acceptable delays in oncologic treatments for patients with a suspected malignancy. The objective of this study was to characterize the association between timing of treatment and survival and to evaluate how extended delays to guideline-concordant treatment for suspected stage I high grade serous ovarian cancer might impact survival. Methods: Using the National Cancer Data Base (2004-2015), patients surgically treated for clinical stage I high grade serous ovarian cancer were identified. Patients receiving surgery within 14 days of diagnosis were compared to those receiving surgery 2-4 months after diagnosis. The groups were propensity matched with 2:1 matching to balance baseline characteristics. Survival outcomes of the two cohorts were assessed using Kaplan-Meier analysis, the log-rank test, and multivariable Cox proportional hazards analysis. Results: During the study period, 10,957 patients underwent guideline-concordant surgery for stage I high-grade serous ovarian cancer within 14 days of diagnosis (early) and 171 patients underwent surgery 2-4 months following diagnosis (delayed). Between patients in the early versus delayed treatment groups, the delayed group was more likely to be Black (5.3% vs 11.8%, p=0.003) and have a smaller tumor size (8.0 vs. 6.2 cm, p<0.001). The early group was more likely to have private insurance (57.4% vs 44.9%, p=0.006). There was no significant difference in 5-year overall survival between the unmatched early versus delayed groups (56.6% vs 50.8%, p=0.36). Propensity-score 2:1 matching was used to create two groups of patients who received early (n=342) or delayed (n=171) surgery that were well-matched with regards to baseline characteristics. The only significant difference between groups in the matched cohort was patients in the delayed group had smaller tumors (8.0 vs 6.2 cm, p<0.001). There was no significant difference in 3-year overall survival (early 73.6% [95% confidence interval (CI): 68.2 - 78.2] vs delayed 72.6% [95% CI: 64.4 - 79.2]) or 5-year overall survival (early 56.6% [95% CI: 50.3 - 62.4] vs delayed 50.8% [95% CI: 41.6 - 59.3], log-rank p=0.52) between the early versus delayed groups (Figure 1). In Cox proportional hazards analysis, age in years (adjusted hazard ratio (aHR)=1.03, [CI: 1.01 - 1.04], p=0.003) and Charlson Deyo comorbidity composite score of 2 (aHR=2.58, [CI: 1.44 - 4.62], p=0.001) were associated with worse survival. Private insurance status was associated with improved overall survival (aHR=0.51, [CI: 0.27 - 0.96], p=0.04). Surgical delay was not significantly associated with different overall survival (aHR=1.19, [CI: 0.90 - 1.58], p=0.23). Conclusions: In this national analysis, extended delay of treatment for stage I high grade serous ovarian cancer was associated with similar survival outcomes when compared to timely treatment in pre-match and post-match cohorts. In the setting of the COVID-19 pandemic, where hospital resources may be limited, delays in surgery are concerning and prompt oncologic treatment is preferred; however, at least based on these retrospective data, delays in surgery for presumed stage I ovarian cancer may not be associated with significantly worse survival.
oncology,obstetrics & gynecology
What problem does this paper attempt to address?