Clinical Validation Of The Molecular Assays For Intraoperative Diagnosis Of Metastases In Sentinel Lymph Node Of Breast Cancer With A Clinicopathologic Study

Axillary lymph node metastasis is the most important prognostic factor of breast cancer patients. Traditionally, surgeons perform axillary lymph node dissection (ALND) to identify the axillary lymph node status to provide staging information and therefore determine the treatment strategy. However, ALND can cause some complications, such as upper limb lymphedema, pain, numbness, impaired shoulder function, and stiffness. Thus, ALND has been gradually replaced by sentinel lymph node biopsy (SLNB). Many studies have shown that a sentinel lymph node (SLN) can accurately predict the status of axillary lymph nodes in early-stage breast cancer patients. If the intraoperative SLNB result is negative, it is highly suggestive that other axillary lymph nodes are not involved, and the patient does not need a further ALND. This alternative will enable the breast cancer patients to avoid the pain caused by the ALND and reduce their financial burden. If the result is positive, an ALND can be immediately performed to avoid a second surgery.As an innovative technique, the application of SLNB in breast cancer still faces some problems, such as the optimal method for intraoperative and postoperative pathological examination of SLN, the pathological evaluation criteria of SLN, whether a routine SLNB should be carried out for patients with ductal carcinoma in situ (DCIS) or ductal carcinoma in situ with microinvasion (DCIS-MI), as well as how to effectively predict the risk of SLN metastases in breast cancers. To illustrate the above-mentioned problems, we undertook a series of research. The current research project is comprised of the following three parts:Part I Evaluation of intraoperative molecular assays for detecting breast cancer metastases in SLNsObjective:(1) To investigate the clinical validation of GeneSearchTM breast lymph node assay (Veridex, ILC, Warren, NJ, USA, GeneSearch for short) for intraoperative diagnosis of metastases in SLNs of breast cancer, and compared the GeneSearch results with intraoperative imprint cytology (ⅡC) and postoperative serial sectioning.(2) To investigate the clinical validation of One-step nucleic amplification assay (OSNA for short), and compared the OSNA results with ⅡC and postoperative serial sectioning.Methods:(1) GeneSearch study:A prospective study of229consecutive SLNs from88patients was conducted in our institution. Every SLN was cut into2mm slabs which were examined by IIC firstly, then alternatively by GeneSearch or postoperative serial sectioning. GeneSearch uses Real-time fluorescence quantitative RT-PCR to detect the expression of CK19and mammaglobin of SLNs. The results of GeneSearch were compared with the results of ⅡC and postoperative serial sectioning.(2) OSNA study:A prospective study of370consecutive SLNs from115patients was conducted in our institution. A total of311SLNs has undergone the OSAN test. All SLNs were sectioned in about2mm pieces. ⅡC was performed on all pieces, and then OSNA and postoperative serial sectioning were performed on alternative node pieces. OSNA uses loop-mediated isothermal amplification technology without the extraction of RNA from genomic DNA and contains six primers to detect the expression of CK19mRNA. The results of OSNA were compared with the results of ⅡC and postoperative serial sectioning.Results:(1) GeneSearch study results:A total of88breast cancer patients with225SLNs (mean2.6SLNs per patient) were enrolled in the GeneSearch study. All patients but one were female, with a median age of60years (range,28-79years). Nineteen patients contained macrometastases. Four patients were diagnosed as micrometastases. Eighty-eight patients were free of metastases; Twenty-seven SLNs were identified as macrometastases. Micrometastases were found in9nodes. The other189SLNs were free of metastases, including5isolated tumor cells (ITCs). The overall rate of agreement between GeneSearch and the postoperative serial sectioning was95.6%(95%confidence interval [CI]:91.7%～97.7%), with a sensitivity of86.1%(95%CI:69.7%～94.8%) and a specificity of97.4%(95%CI:93.6%～99.0%) based on the number of SLNs. And the positive predictive value (PPV) and negative predictive value (NPV) of GeneSearch were of86.1%(95%CI:69.7%～94.8%) and97.4%(95%CI:93.6%～99.0%) respectively. Positive SLN metastasis size was correlated with the RT-PCR cycle threshold (Ct value).(2) OSNA study results:A total of115patients with370SLNs (mean3SLNs per patient) were enrolled in the OSNA study. All patients were female, with a median age of50years (range,24～80years). Nineteen patients contained macrometastases. Seven patients harbored micrometastases. Eighty-nine patients were free of metastases; Twenty-five of the370SLNs were identified as macrometastases. Micrometastases were found in11nodes. Four SLNs harbored ITCs. The other271SLNs were free of metastases. Fifty-nine nodes were too small to be sampled for the OSNA test. The overall rate of agreement between OSNA assay and the postoperative serial sectioning was95.2%(95%confidence interval [CI]:91.6～96.9%), with a sensitivity of83.3%(95%CI:66.5～93.0%), a specificity of96.7%(95%CI:93.7～98.4%), a PPV of76.9%(95%CI:60.3～88.3%), a NPV of97.8%(95%CI:95.0～99.1%) based on the number of SLNs sampled. Eleven out of15discordant SLNs can be explained by "tissue allocation bias".Conclusions:(1) GeneSearch assay is standardized, objective, and reproducible and can utilize more lymphoid tissue. The performance of GeneSearch test for intraoperative diagnosis of SLNs is equivalent to the postoperative serial sectioning. GeneSearch can obtain a relatively higher sensitivity than IIC, especially for the micrometastases SLNs. We recommend GeneSearch be used in daily clinical diagnostic work.(2) OSNA assay is also standardized, objective, and reproducible and can utilize more lymphoid tissue. In addition, it can distinguish between micrometastasis and macrometastasis to enable us to make further study of the significance of micrometastasis. The performance of OSNA assay for intraoperative diagnosis of SLNs is equivalent to the postoperative serial sectioning. We recommend that OSNA assay can be used in daily clinical diagnostic work.(3) Although GeneSearch and OSNA have their own advantages and disadvantages, both of them can provide a good performance; and therefore can be good alternative methods for traditional pathological diagnosis.Part II SLNB in Patients with DCIS and DCIS-MIObjective:The role of SLNB in DCIS or DCIS-MI is still in controversy. The purpose of our study is to determine if there is a need for SLNB in patients with pure DCIS or DCIS-MI.Methods:One hundred and forty five patients with DCIS or DCIS-MI who underwent SLNB between March2009and March2011at the Fudan University Shanghai Cancer Center (FUSCC) were enrolled in our cohort. There were85patients with DCIS and60patients with DCIS-MI,20patients (8DCIS,12DCIS-MI) of whom were also tested by the GeneSearch Assay. The clinicopathological parameters such as the tumor size, histological grade, clinical presentation, mammographic manifestation, necrosis, and the status of ER and PR were summarized. The correlation between these parameters and SLN metastasis was analyzed. We also evaluated whether SLNB is required in patients with DCIS or DCIS-MI.Results:(1) Clinical features of DCIS patients:All patients were female. The ages ranged from23to84years (mean50years, median51years). The tumor size ranged from0to6cm, with a mean of1.4cm. A total of241SLNs was successfully detected with an average of3SLNs per patient. Only one (1.2%) patient showed a micrometastatic SLN. Twelve patients also underwent an ALND. None was demonstrated a metastasis. Statistical analysis showed no identified risk factors of SLN metastasis.(2) Clinical features of DCIS-MI patients:All patients were female. The ages ranged from30to81years (mean52years, median53years). The tumor size ranged from0to5.5cm, with a mean of2.4cm. A total of171SLNs was successfully detected with an average of3per patient. Two patients (2/60,3.3%) were detected to have macrometastatic SLNs. In addition, two cases displayed ITCs. Twelve patients also undergone ALND and none of them showed a positive lymph node. Statistical analysis displayed no risk factors of SLN metastasis.(3) GeneSearch Assay results:None of the8DCIS patients showed a positive result. In12DCIS-MIs,2patients were positive. The results were fully consistent with the results of serial sectioning.Conclusions:(1) Based on our currently available data, the routine use of SLNB in all patients with DCIS or DCIS-MI is not warranted. It is necessary to select certain kind of patients initially diagnosed with DCIS to undergo SLNB. However, it is also acknowledged that there are still no definite risk factors used to indicate an invasive component will be found after surgery.(2) Selected DCIS-MI patients with large tumor, high grade, several invasive clusters and vascular invasion seem to need a concomitant SLNB. Nevertheless, this should be certified by further investigation with larger numbers of patients and long follow-up.(3) Most of the SLN-positive DCIS or DCIS-MI patients were micrometastases or ITCs whose clinical significance and prognostic implications is still in controversy. Part Ⅲ Expressions and significance of aB-crystallin and Galectin-7in breast cancer with different metastatic types of SLNObjective:The purpose of the study is to use aB-crystallin and Galectin-7to detect the expression in breast cancers with different metastatic types of SLN, and to explore their significance.Methods:One hundred and seventy four patients with IDC not otherwise specified who underwent SLNB between2005and March2009at FUSCC were enrolled in our cohort. The paraffin-embedded tissue samples of the primary breast cancer, the corresponding normal breast and the SLNs of macrometastatic cases were collected and included in tissue microarrays (TMAs), and stained with antibodies against aB-crystallin and Galectin-7.Results:(1) All patients were female. The ages ranged from28to85years (mean51years, median51years). The tumor size ranged from0to5.5cm, with a mean of2.2cm. A total of484SLNs was successfully detected with an average of3SLNs per patient. Among these174patients,96cases were diagnosed as macrometastases,23cases were micrometastases, and the other55cases were free of metastasis (including5ITCs).(2) Expression of aB-crystallin:Fifty-nine out of the96(61.4%) macrometastatic cases were positive for aB-crystallin. Eight out of the23(34.8%) micrometastatic cases were positive for aB-crystallin. Twelve out of the55(21.8%) SLN-negative cases were positive for aB-crystallin. In the metastasis group (including macrometastasis and micrometastasis), the expression of aB-crystallin was significantly higher than the non-metastasis group{χ2=18.045, P=0.000). The expression of aB-crystallin in macrometastasis group was statistically higher than the non-metastasis group (χ2=22.056, P=0.000). The SLN-metastatic tumors showed a significantly increased aB-crystallin protein-positive rate than the primary tumors (86.8%vs61.4%, χ2=12.631, P=0.000). Our study also revealed that aB-crystallin protein levels were gradually and significantly elevated with the increase of histological grade (χ2=7.306, P=0.026).(3) Expression of Galectin-7:Although there was no obvious difference in the cytoplasmic Galectin-7expressions among the three different metastasis groups (χ2=2.398,P=0.301), a significantly different nuclear-positive rate can be observed. In the metastasis group (including macrometastasis and micrometastasis), the nuclear-positive rate of Galectin was significantly lower than the non-metastasis group (χ2=18.045, P=0.000). The nuclear expression of Galectin-7in macrometastasis group was statistically lower than the non-metastasis group (χ2=6.959, P=0.008). The SLN-metastatic tumors showed a significantly decreased Galectin-7nuclear-positive rate than the primary tumors (17.6%vs45.8%, χ2=14.063, P=0.000), while the SLN-metastatic tumors showed a significantly increased Galectin-7cytoplasm-positive rate than the primary tumors (strong positive cases:54.4%vs35.4%, χ2=9.433, P=0.007). The nuclear expression of Galectin-7levels were gradually and significantly decreased with the increase of histological grade (χ2=16.732, P=0.000). In HER2overexpressing and triple-negative groups, the nuclear-positive rate of Galectin-7is lower than luminal A or luminal B group.Conclusions:The expression of aB-crystallin may play an active role in the metastasis of breast cancer, and might be served as an important parameter for determining tumor biological behavior. The early nuclear expression of Galectin-7may be a negative regulatory factor in breast cancer development, while the gradual disappearance of Galectin-7nuclear expression is likely to be a sign of breast cancer progression.