| Objective As a first-echelon nodal drainage site in breast cancer,the status of axillary lymph nodes(ALN)and internal mammary lymph nodes(IMLN)is valuable for regional staging,prognosis evaluation and treatment choice.At present,the diagnosis and treatment of ALN is close to the individualized level,but the best management of IMLN is still controversial.With the development of sentinel lymph node biopsy in breast cancer,the status of ALN has been well assessed with the axillary sentinel lymph node biopsy(A-SLNB)procedure and individualized management could be performed,but the technology of internal mammary sentinel lymph node biopsy(IM-SLNB)lags far behind that of A-SLNB.Although the"Clinical practice Guide for NCCN Breast Cancer" has long incorporated the internal breast sentinel lymph node(IMSLN)concept,routine performance of the IM-SLNB in breast cancer patients remains a subject of debate due to the low IMSLN visualization rate and unclear clinical relevance.On the one hand,the traditional tracer injection technique was only able to identify IMSLN in a small proportion of patients(visualization rate 13%;range from 0%to 37%),which has been the restriction for the IM-SLNB to date.On the other hand,the IM-SLNB indications have not been standardized in current guidelines.Clinical work/study still referring to the indications of A-SLNB and only performing in clinically ALN-negative patients,which greatly reduces the clinical benefit of IM-SLNB.In this prospective multicenter clinical study,the modified tracer technique was performed during the IM-SLNB procedure to verify whether it can achieve the ideal IMSLN visualization rate,and to explore the clinical significance of IM-SLNB in different populations(especially clinically ALN-positive patients).In order to solve the technical bottleneck of IM-SLNB and standardize its indications,and to further explore the value of IM-SLNB in individualized treatment of IMLN.Methods A total of 591 patients with primary breast cancer(cT1~3N0~2M0)were prospectively enrolled in 7 centers in China from August 2018 to February 2020,the median age was 49(26~70).The study protocol was approved by the ethics committee of each center and registered as "Clinicaltrials.gov"(ID:NCT03541278).All patients were injected radionuclide tracer(99mTc-sulfur colloid)into the parenchyma of breast gland by modified tracer technique(double quadrant/periareolar area,high volume,and ultrasound guidance)2 to 20 hours before operation,and some patients underwent SPECT or SPECT/CT lymphoscintigraphy(LSG)pre-operation.The IMSLN was visualized and located by gamma detector after breast and axillary operations,then IM-SLNB was performed via intercostal approach.Routine H&E staining pathological examination was required for IMSLN,and CK-19 immunohistochemical detection was performed for patients with negative H&E staining to exclude micrometastases.The staging was according to the 8th edition of AJCC breast cancer staging criteria,and the adjuvant treatment was reference to the latest version of NCCN guidelines and St.Gallen expert consensus.The IMSLN visualization rate and related factors,IMSLN distribution,IM-SLNB related technical indicators,IMSLN metastasis rate and predictive factors were analyzed,and a nomogram(model)was constructed to predict the IMSLN status.The effects of IM-SLNB in different populations(clinically ALN-negative and clinically ALN-positive patients)on staging and treatment decision were compared and analyzed.Results The overall visualization rate of IMSLN was 66.0%(390/591),and that of each centers was 60.7%,63.6%,84.0%,73.9%,77.3%,50%and 88.2%,respectively.The visualization rate of IMSLN was significantly correlated with body mass index(P=0.022)and time of radionuclide injection to visualization(P<0.001),but not with age,tumor size,tumor location,pathological type,histological grade and radionuclide intensity(all P>0.05).Controlling the time of radionuclide injection to visualization within 1 half-life(6h)could significantly improve the IMSLN visualization rate(75.2%vs.48.5%,P<0.001).After 20 cases of technical learning,radionuclide injectors could skillfully master the "modified tracer technique" and obtain a higher IMSLN visualization rate(>70%).The overall success rate of IM-SLNB was 97.2%(379/390),and that of each centers was 96.5%,98.4%,98.5%,100%,100%,77.8%and 100%,respectively.The median time of IM-SLNB procedure was 7mins(3-25mins).Intraoperative complications included extra skin incision(2.6%,10/379),internal breast vascular injury(3.2%,12/379)and parietal pleural injury(6.9%,26/379).And the operator could skillfully master IM-SLNB after completing the learning curve of 40 cases.Only 1 patient(0.3%)developed postoperative pneumothorax.IMSLN were detected from the 1st to 4th intercostal space(ICS)(accounting for 33.9%,36.0%,24.9%,5.2%,respectively),with the median number of IMSLN removed was 2 per person(total 667;range,1-8 IMSLN per person).Whether the patients underwent LSG or not did not affect the IMSLN visualization rate(68.7%vs.63.8%,P=0.213),the IM-SLNB success rate(96.7%vs.97.6%,P=0.620)and the IMSLN number(P=0.251).The overall IMSLN metastasis rate in patients who underwent IM-SLNB was 20.1%(76/379),and clinically ALN-positive patients were significantly higher than clinically ALN-negative patients(37.7%vs.11.7%,P<0.001).A total of 104 positive IMSLN were removed(95 macrometastases and 9 micrometastases),with a median number was 1 per person(range,1~7 per person),which were distributed in the first ICS 37.5%,the second ICS 35.6%,the third ICS 21.2%and the fourth ICS 5.8%.Univariate analysis showed that IMSLN metastasis was significantly correlated with tumor size,tumor location,histological grade and number of positive ALN,but not with age,number of IMSLN,pathological type and molecular subtype.Multivariate analysis showed that tumor size(P=0.028),tunor location(P<0.001)and number of positive ALN(P<0.001)were independent predictors of IMSLN metastasis.Those variables were included in nomogram,whose predictive ability was better than that of ALN positive number(area under the curve:0.860 vs.0.804,P<0.001).IM-SLNB led to a more accurate nodal category in all the 379 patients who underwent IM-SLNB successfully and to more advanced staging in IMSLN positive patients.In the overall population,IM-SLNB could optimize the lymphatic staging of 20.1%(76/379)patients and the systemic adjuvant therapy of 1.6%(6/379)patients,and provide important radiotherapy reference information for 52.2%(198/379)patients.In the clinically ALN-negative subgroup,the lymphatic staging of 30 patients(11.7%)could be advance,the systemic adjuvant therapy of 6 patients(2.3%)could be optimize,and the radiotherapy decisions of 76 patients(29.6%)could be potentially affected.In the clinically ALN-positive subgroup,the lymphatic staging of 46 patients(37.7%)could be advance,the systemic adjuvant therapy did not change,and the radiotherapy decisions of 122 patients(100%)could be potentially affected.Conclusions As a minimally invasive technique,IM-SLNB guided by modified tracer technique can diagnose IMLN status and provide accurate lymphatic staging and adjuvant therapy strategies,which are mainly reflected in three aspects:(1)Technical feasibility:the modified tracer technique has a high IMSLN visualization rate,and it has good repeatability through the prospective national multicenter study.IM-SLNB is a minimally invasive staging technique for breast cancer with high success rate,easy to master and few complications.(2)Clinical indications:IM-SLNB should be performed not only in clinically ALN-negative patients,but also in clinically ALN-positive patients.We recommend that IM-SLNB be routinely performed during mastectomy and selectively performed during breast-conserving surgery(medial tumor and>2cm/ALN positive).(3)Accurate staging and individualized therapy:IM-SLNB can obtain the histological diagnosis of IMLN and provide more accurate regional lymphatic staging for breast cancer patients.Although the adjuvant systemic therapy strategies were rarely affected,important reference information could be provided for radiotherapy decision-making. |