TL;DR
This study reveals that the number of positive lymph nodes (PLNs) is a more powerful predictor of survival in oral tongue squamous cell carcinoma (OTSCC) than current staging methods. By proposing a simplified nodal category (0, 1-2, and ≥3 PLNs), the researchers offer a more accurate way to stratify patients and guide treatment decisions, including the use of adjuvant chemoradiotherapy.
What Problem Does This Paper Address?
Oral tongue squamous cell carcinoma (OTSCC) is a challenging cancer with a significant impact on survival, often due to the spread of cancer to nearby lymph nodes. Current staging systems, like the AJCC's pN staging, aim to predict patient outcomes, but they have limitations. These include complex categories, overlapping survival curves, and a one-size-fits-all approach that may not adequately capture the nuances of OTSCC's biological behavior. This can lead to suboptimal risk stratification and treatment planning. The paper seeks to address this by identifying a more precise and practical metric for predicting survival and guiding treatment in OTSCC patients.
Methods
The researchers conducted a retrospective study using data from two sources: an institutional registry of 518 OTSCC patients who underwent curative resection and the Surveillance, Epidemiology, and End Results (SEER) database for external validation.
Key methodological aspects include:
- Multivariate Cox Proportional Hazards Models: Used to identify independent predictors of survival and calculate hazard ratios (HRs) and confidence intervals (CIs) for various pathological nodal features.
- Propensity Score Matching (PSM): Employed to minimize selection bias and create balanced groups for comparing the effectiveness of adjuvant chemoradiotherapy (ACRT) in patients with different nodal burdens. This helps ensure that observed differences in outcomes are attributable to the treatment rather than pre-existing patient characteristics.
- Kaplan-Meier Analysis: Used to estimate survival probabilities and compare survival curves between different patient groups.
- Development of a Simplified Nodal Category: Based on the number of positive lymph nodes (PLNs), the authors proposed a new categorization (0 PLN, 1-2 PLNs, and ≥3 PLNs) to improve prognostic accuracy.
- C-index Evaluation: Used to assess and compare the predictive ability of the proposed nodal category against the existing AJCC pN-staging system.
Datasets
- Institutional Registry: Data from 518 consecutive adult patients with newly diagnosed, nonmetastatic OTSCC who underwent curative surgical treatment between January 2010 and December 2020.
- SEER Database: Data from 13,412 OTSCC patients (with exclusions leading to a final validation cohort of 2573 patients) who underwent curative resection without distant metastasis, collected between 2010 and 2015.
Key Insights and Ideas
- Number of PLNs Reigns Supreme: The study strongly suggests that the number of positive lymph nodes is a more significant predictor of survival in OTSCC than other established nodal features like extranodal extension (ENE) or lower neck involvement. This implies that the sheer burden of metastatic disease in the lymph nodes is a critical determinant of prognosis.
- A Simpler Nodal Staging is Better: The current AJCC pN staging system, while comprehensive, can be complex. The proposed simplified nodal category (0, 1-2, and ≥3 PLNs) demonstrates improved predictive accuracy, suggesting that a more streamlined approach can be more effective for risk stratification.
- Threshold for Adjuvant Therapy: The research identifies a critical threshold of ≥3 positive lymph nodes as a key indicator for patients who may benefit most from adjuvant chemoradiotherapy (ACRT). This has direct implications for treatment planning, potentially sparing some patients unnecessary toxicity while ensuring others receive potentially life-saving treatment.
- Downstaging Potential: The similar prognoses observed between patients with one and two PLNs led to a proposal for downstaging current pN2 patients with 1-2 PLNs to pN1. This refinement could lead to more accurate risk assessment and potentially better treatment allocation.
Key Results
- Predictive Power of PLN Count: In multivariate analysis, the number of PLNs independently predicted overall survival (OS), outperforming other nodal features.
- Survival Differences by PLN Count:
- Patients with 0 PLNs had a median OS of 40.2 months.
- Patients with 1-2 PLNs had a median OS of 32.1 months, significantly worse than those with 0 PLNs.
- Patients with ≥3 PLNs had the poorest prognosis with a median OS of 19.0 months.
- Proposed Nodal Category Performance: The simplified nodal category (0, 1-2, ≥3 PLNs) showed improved predictive ability (C-index: 0.717) compared to the 8th edition AJCC pN-staging system (C-index: 0.713) in the institutional cohort, and this was validated in the SEER registry.
- ACRT Benefit: Propensity score matching revealed that ACRT significantly improved OS in patients with advanced nodal disease (≥3 PLNs), with a 5-year OS of 55.3% for ACRT vs. 26.2% for non-ACRT in the institutional cohort, and 37.1% vs. 24.7% in the SEER registry. No significant benefit was observed for patients with 0 or 1-2 PLNs.
Limitations and Future Work
- Nodal Size Data: The institutional registry lacked data on nodal size, a parameter that has shown some prognostic value in previous studies. This limits a complete comparison with staging systems that incorporate nodal size.
- Replacement of Current Staging: The proposed nodal scheme represents a significant adjustment to current practice, and its full integration would require substantial effort. The authors emphasize their aim is to optimize, rather than replace, the existing system.
- Heterogeneity in pN0 Group: The pN0 group, which constituted over half of the study population, exhibits heterogeneous prognoses. The authors suggest that further research is needed to identify additional biomarkers to refine predictions within this group.
- Future Directions: The study highlights the need for further exploration into the underlying mechanisms that make the number of PLNs a superior prognostic factor compared to ENE. Additionally, the authors advocate for the consideration of their simplified nodal classification in future updates to staging systems for oral cavity cancers.
Summary updated