Critical appraisal:Berger AC, Fierro M, Kairys JC, Berd D, Sato T, Andrel J, et al 2012
Critical Appraisal
Berger AC, Fierro M, Kairys JC, Berd D, Sato T, Andrel J, et al. Lymph node ratio is an important and independent prognostic factor for patients with stage III melanoma. J Surg Oncol 2012 Jan;105(1):15-20 Available from: http://www.ncbi.nlm.nih.gov/pubmed/21815149.
Key Facts
case series
Hypothesized that the ratio of metastatic to examined lymph node ratio (LNR) would be the most important prognostic factor for stage III patients.
168
Survival
Median survival: 34 months (95% CI 29–44) with 36% (95% CI 28–44%) of patients surviving at least 5 years. Using univariate analysis predictors of overall survival (OS) was LNR >25% (HR = 3.08, P < 0.01). A LNR between 10% and 25% was also an important predictor of survival. Other factors that were significant: type of treatment, number of positive nodes, and LN basin dissected (patients who had LND for cervical metastases had decreased survival compared to those with axillary metastases.) Total number of examined or excised lymph nodes was not an important factor for survival (HR = 1.0 per node, P = 0.83). Borderline significant difference in survival depending on whether the indication for LND was performed for a positive sentinel node compared to those who underwent dissection for a clinically positive lymph node basin (P = 0.08), as well as in primary sites with lower hazards in patients with either trunk or extremity versus head/neck (P = 0.10, 0.09, respectively).
Multivariate analysis: only factors predictive of overall survival were LNR and type of treatment. LNR >25% had a fourfold higher risk of dying compared to those with a LNR <10% (P < 0.01). Type of treatment: individual levels show of the 4 types compared to none/unknown, only chemotherapy marginally significant hazard ratio (HR = 1.90, P = 0.08). LN basin was not a globally significant factor in the model (P = 0.20), >1 nodal basin involved had twice the risk of dying (HR = 2.12, P = 0.05) compared to axillary. When included in the model with LNR, as an alternate analysis, the total number of positive nodes was not significant, P = 0.75, although this is somewhat obscured by the fact that this value is used to compute (and may be collinear with) LNR. LNR was still significant, P < 0.024.
Median and 5-year survival patients by LNR, significant difference among the 3 LNR groups. Group 1: median survival of 77 months (95% CI 43, ∞) estimated 5-year survival 52% (95% CI 40–63%). Groups 2 and 3 had 5-year survivals of 24% and 0%. The separation of the 3 survival curves was persistent throughout the entire course of follow-up. Median and overall survival based on number of positive nodes (AJCC N stage N1, N2, and N3). Difference in survival: N1 median of 52 months, 26 months for N3. Survival curves of N1 and N2 patients do not have a lot of separation from each other. LNR model provided a better fit for overall survival than AJCC N stage.
no
Evidence ratings
IV
| High risk of bias | Comments: Please replace this text and include any additional comments in regards to your risk of bias rating |
| 1 | Additional comments: Please replace this text and briefly describe the reasons for your rating |
Included
This appraisal has been completed.
- Article
- Berger AC, Fierro M, Kairys JC, Berd D, Sato T, Andrel J, et al. Lymph node ratio is an important and independent prognostic factor for patients with stage III melanoma. J Surg Oncol 2012 Jan;105(1):15-20 Available from: http://www.ncbi.nlm.nih.gov/pubmed/21815149.
- Assigned to
- User:Jackie.buck
- Topic area
- Guidelines:Property "Appraisal topic" (as page type) with input value "Guidelines:" contains invalid characters or is incomplete and therefore can cause unexpected results during a query or annotation process.
- Clinical question
Section below only relevant for Cancer Council Project Officer