Intra-operative Molecular Diagnosis of Sentinel Lymph Node and Prediction of Non-sentinel Lymph Node Metastasis in Breast Cancer Patients
Sentinel lymph node biopsy (SLNB) has become a standard procedure for node-negative breast cancer patients. The accurate intra-operative assessment of the sentinel lymph node (SLN) is crucial as it enables axillary lymph node dissection (ALND) to be performed synchronously during breast surgery. This avoids the morbidity, inconvenience, and additional costs associated with a second operation. One-step nucleic acid amplification (OSNA) assay, developed by Sysmex in Kobe, Japan, is a molecular technique that combines lymph node tissue homogenization with reverse-transcription loop-mediated isothermal amplification of cytokeratin-19 (CK-19) mRNA in a single reaction. This assay can accurately detect nodal metastases larger than 0.2 mm intra-operatively.
Two studies were conducted to validate the intra-operative OSNA assay, registered under Clinical trial registry Nos. CBCSG-001c and NCT03937414. The first trial, referred to as “trial 1,” involved five centers and included 552 patients. Conducted in 2010, this trial confirmed the good performance of the OSNA assay, with an accuracy of 91.4% and a sensitivity of 83.7%. The second trial, “trial 2,” involved three centers and included 1090 patients, conducted from June 2015 to May 2017. The results demonstrated that the sensitivity, specificity, and accuracy of the OSNA assay in SLN-positive breast cancer patients were 88.72%, 91.28%, and 90.83%, respectively.
The findings from these two large randomized trials have significantly influenced clinical practice for breast cancer patients with SLN metastases. The ACOSOG Z0011 trial demonstrated that patients with T1-2 tumors who received breast-conserving treatment could be exempt from ALND and axillary radiotherapy when the number of metastatic SLNs is only 1-2. The AMAROS trial suggested that if axillary treatment is necessary for SLN-positive patients, axillary radiotherapy is preferable to ALND, as it provides similar local control with less morbidity. Therefore, for patients who do not meet the criteria of the Z0011 and AMAROS trials, ALND remains the standard treatment. Additionally, for SLN-positive patients undergoing mastectomy who meet the AMAROS trial criteria, ALND is one of the alternatives.
However, it has been reported that 20% to 60% of patients with positive SLNs do not develop non-sentinel lymph node (NSLN) metastasis, making ALND an over-treatment for these patients. Consequently, there is a significant clinical need for a predictive nomogram model to distinguish SLN-positive patients who do not require ALND. Such a model would be invaluable for guiding subsequent surgical treatment.
In trial 1, data from 103 patients who underwent ALND due to positive SLNs were collected to construct the nomogram. Variables selected from traditional predictive models, such as Memorial Sloan-Kettering Cancer Center (MSKCC), MD Anderson (MDA), Mayo, Tenon, Cambridge, Stanford, Helsinki, and total tumor load (TTL, the sum of the CK-19 mRNA copy number/mL of all positive lymph nodes), were adopted as clinicopathological indicators to construct the model. Logistic multivariate regression analysis was performed on the statistically significant variables, and a novel nomogram model for predicting NSLN metastasis was established using TTL, clinical primary tumor size, and the number of positive and negative SLNs. The area under the curve (AUC) of the receiver operating characteristic (ROC) analysis of the model in trial 1 was 0.814. In trial 2, data from 159 patients who underwent ALND due to positive SLNs were used to validate the nomogram. The AUC of the ROC analysis of the model in trial 2 was 0.842, with a sensitivity and specificity of 81.48% and 79.41%, respectively. A comparative analysis of the model with traditional predictive models (represented by MDA and Tenon models) was conducted on the data from trial 2. The AUCs of MDA and Tenon were 0.745 and 0.623, respectively. It was evident that the AUCs of the new nomogram were statistically superior to those of the MDA and Tenon models.
The 2016 breast cancer National Comprehensive Cancer Network guideline recommended irradiation of the internal mammary for patients with more than four positive axillary lymph nodes (category 1) and strongly considered internal mammary radiation for patients with one to three positive axillary lymph nodes (category 2A). Consequently, radiotherapists have also been focusing on the value of predictive models. Because the local regional treatment of patients with axillary lymph node status (pN) of pN1 and ≥pN2 are distinct, the predictive model may help radiotherapists delineate the radiotherapy target more accurately. To explore the predictive power of the nomogram in patients with one to three metastases and ≥4 metastases in axillary lymph nodes, combined data from 262 patients from trial 1 and trial 2, including 193 with pN1 and 69 with ≥pN2, were analyzed. The cut-off value of the new model, which was sensitive to discern patients with pN1 and ≥pN2, was 0.454, and the AUC was 0.861.
Researchers worldwide are striving to develop models for predicting the latent risks of NSLN metastasis. Existing models include MSKCC, MDA, Mayo, Tenon, Cambridge, Stanford, Helsinki, and others, some of which have already been applied in clinical practice. However, traditional predictive models have several defects. First, although histological evaluations of SLNs are commonly used in all models, no unified standard has been confirmed by various institutions to minimize discrepancies. Second, it is difficult to precisely measure the maximum size of a complex three-dimensional metastasis in SLNs merely by the conventional histological evaluation of two-dimensional sections. Third, information for the construction of traditional models can only be achieved post-surgically, including tumor grade, multi-focus, and vascular invasion, which means that it is impossible to adopt a traditional model pre-operatively or even to guide an axillary operation. These defects significantly constrain the clinical application of traditional predictive models.
In contrast, the new model established on the basis of intraoperative molecular diagnostics is objective and standardized. Information on primary clinical tumor size, TTL, and the number of positive and negative SLNs can be obtained pre-operatively and intra-operatively. Unlike conventional models that can only predict based on post-operative information, the new model allows for rapid intra-operative prediction to guide subsequent axillary treatment timely. Moreover, the new model can simultaneously distinguish the risk of lymph node metastasis in breast cancer patients with pN1 and ≥pN2, which is helpful to pinpoint the target area of radiotherapy in clinical practice.
In summary, the OSNA assay is an accurate intraoperative assessment for breast SLNs. The nomogram for predicting NSLN metastasis based on TTL, primary clinical tumor size, and the number of positive and negative SLNs presents superior performance over other predictive models. The novel model will help guide axillary management and precisely confirm the target region of radiotherapy.
doi.org/10.1097/CM9.0000000000000609
Was this helpful?
0 / 0