Jpn J Clin Oncol 2002;32(10)403–406
Clinically Useful Detection Criteria for Sentinel Nodes in Patients with
Breast Cancer Using a Radioisotope Technique
Kazuhiko Sato1, Kuniyoshi Tamaki1, Takashi Shigekawa1, Hitoshi Tsuda2, Shigeru Kosuda3, Shoichi Kusano3,
Hoshio Hiraide4 and Hidetaka Mochizuki1
1Department
of Surgery I, 2Department of Pathology II, 3Department of Radiology and 4Research Institute, National
Defense Medical College, Tokorozawa, Saitama, Japan
Received February 28, 2002; accepted June 20, 2002
Background: The radioisotope technique has been used to identify sentinel nodes in patients
with breast cancer. However, quantitative analysis of the radioactivity for detecting the sentinel
nodes was not previously examined. In this study, we considered a clinically useful criterion for
detecting sentinel nodes by a detailed analysis of 312 sentinel nodes using the radioisotope
technique.
Patients and methods: Patients with T1–2, N0 breast cancer were eligible for this study. The
nodes with the highest radioactivity after injection of technetium-labeled tin colloids were
identified as hot nodes. The radioactivities of the hot nodes and the background counts of the
axillary basin were examined in order to establish new criteria for detecting the sentinel nodes.
Results: Between May 1997 and December 2001, 312 hot nodes were detected in 183 of 186
patients (98.4%). Since the false-negative rate for metastasis in hot nodes was only 2.1%
(1/48), they could serve as sentinel nodes to predict the nodal status. However, there was a
wide distribution of the hot nodes and the background in terms of absolute counts and a criterion for the sentinel nodes could not be established in terms of the absolute counts. When we
adopted the criterion of sentinel nodes with a ³100 count ratio in relation to the background,
only 169 hot nodes (54.3%) met our definition. When the criterion of a ³10 count ratio was
adopted, all hot nodes met our definition and all other nodes remained non-sentinel nodes.
Conclusion: The criterion for defining sentinel nodes in our method is a node with a ³10 count
ratio with respect to the background. It is recommended that an analysis based on such objective data should be investigated in order to provide surgeons with more accurate and clinically
useful criteria for detecting sentinel nodes.
Key words: detection criteria – sentinel node – breast cancer – radioisotopes
INTRODUCTION
The sentinel node (SN) has been defined as the first lymph
node on a direct drainage pathway from the primary tumor site
and is used to predict the status of the remaining nodes with
regard to metastasis. The SN concept was popularized by
Morton et al. (1) in melanoma and extended to breast cancer
by Krag et al. (2) and Giuliano et al. (3). Various techniques
are used for identifying SNs: using blue dye alone, radiocolloid
alone or both combined. No matter which technique is used,
sentinel node biopsy (SNB) has become widely accepted as a
method of staging the regional lymph nodes in breast cancer
patients (4). Among these methods, the radioisotope technique
has frequently been used together with SNB in the majority of
institutes not only as an investigational procedure but also in
clinical practice (5). However, the level of radioactivity necessary to define the SN has not previously been studied and there
is controversy regarding its definition among institutions.
In this study, to make the identification of SNs easier and
more accurate, we made a detailed analysis of 312 SNs identified at the National Defense Medical College Hospital in order
to investigate the criteria for SN determination.
PATIENTS AND METHODS
For reprints and all correspondence: Kazuhiko Sato, Department of Surgery I,
National Defense Medical College, 3–2 Namiki, Tokorozawa, Saitama 359–
8513, Japan. E-mail: [email protected]
Patients with clinical stage T1–2, N0 breast cancer were eligible for participation between May 1997 and December 2001.
Exclusion criteria for patients were pregnancy, the presence of
multiple primary breast tumors and the clinical suspicion of or
© 2002 Foundation for Promotion of Cancer Research
404
Detection criteria for SN using RI
Table 1. Clinicopathological features of the patients
Characteristic
No.*
Age (years)
59.5 (36–83)
<40
46
40–64
103
³65
34
Body-mass index
23.3 (16.2–30.5)
<25
141
³25
42
Clinical tumor size
T1
94
T2
89
Pathological tumor size (cm)
Figure 1. Numerical distribution of hot nodes.
2.6 (0.2–9)
<3
123
³3
60
Tumor location
Upper quadrants
148
Lower quadrants
35
Prior surgery
Tumor intact
162
Open biopsy
21
Pathological node status
Node-negative
128
Node-positive
55
*The number of patients whose hot nodes were identified.
overt presence of abnormal axillary nodes as determined by
ultrasonography. The study was reviewed and approved by the
Institutional Review Board at the National Defense Medical
College and informed consent was obtained from all patients.
The identification of radioactive nodes (hot nodes) as candidates for SNs was carried out using radioactive tin colloids
(Nihon Mediphysics, Tokyo, Japan), as reported previously
(6–8). In brief, 1 or 3 ml of technetium-99m-labeled tin colloid
(74 MBq/ml) was injected at three sites around the tumor or
biopsy cavity under ultrasonographic guidance with or without
subdermal injection, 2 h prior to surgery. The breast was manually compressed and gently massaged for 1 min. Just before
operation, a concomitant dye injection of 5 ml of indigocarmine (Daiichi Pharmaceutical, Tokyo, Japan) at approximately
the same location as the radiocolloid was performed to facilitate the passage of the radiocolloid into the lymphatics by an
increase in interstitial pressure (7). Again, a 1 min massage
was performed. The nodes with the highest radioactivity were
excised as hot nodes and their ex vivo radioactivities (counts
per second; cps) were determined using a hand-held gammadetector probe (Navigator Systems, USSC, Norwalk, CT).
Background counts were also obtained from the axillary nodal
basin after the removal of the injection site in order to diminish
the artifacts from the primary site. When the background did
not show any signal, its radioactivity count was assumed to be
1 cps in subsequent calculations.
All hot nodes were bisected from the hilum to the periphery.
Each face of the cross-section of the specimen was processed
by frozen-section examination using conventional hematoxylin–eosin staining; the remaining portion of the lymph node
was embedded in paraffin and sectioned for re-examination by
hematoxylin–eosin staining. Since May 1999, if the results of
pathological examination of the hot nodes were positive for
metastases by frozen section, standard level I and II lymph
node dissections were performed in patients with T1N0 breast
cancer.
The hot nodes for the remaining axillary nodes were examined in order to evaluate the predictive value of hot nodes as
SNs. The false-negative rate for metastasis was defined as the
number of patients with negative hot nodes and positive nonhot axillary nodes divided by the number of patients with
positive axillary nodes.
Second, the distributions of absolute counts and count ratios
of the hot nodes and their backgrounds were examined in order
to clarify the useful criteria for the detection of the SN using
the radioisotope technique. The radioactivity count ratios were
determined from the ex vivo hot nodes and the post-excision
lymph node basin.
RESULTS
Between May 1997 and December 2001, a total of 186 women
was enrolled in the study. The mean age was 53.5 years (range
28–83 years). The patients’ characteristics are described in
Table 1. Of the 186 patients, hot nodes were identified in 183,
i.e. an identification rate of 98.4%. A total of 312 hot nodes
(mean number of hot nodes, 1.7) were detected. In 72 patients
(39%), more than one hot node was identified (38 patients, two
hot spots; 17 patients, three hot spots; 17 patients, four or more
hot spots) (Fig. 1).
Table 2 presents the status of the lymph nodes in all procedures followed by axillary lymph node dissection. Fifty-eight
patients were excluded from the false-negative analysis owing
Jpn J Clin Oncol 2002;32(10)
405
Table 2. Cross-tabulation of the number of positive and negative hot nodes
against axillary involvement with metastases
Hot node
Axillary node status
Positive
Negative
Total
Positive
47
0
47
Negative
1
77
78
48
77
125
Total
False-negative rate, 2.1% (1 of 48); sensitivity, 97.9% (47 of 48); specificity,
100% (77 of 77); negative predictive value, 98.7% (77 of 78); overall accuracy, 99.2% (124 of 125).
to a lack of axillary lymph node dissection. The false-negative
rate was 2.1% (1/48). The accuracy of the hot nodes in detecting metastatic disease was 99.2% and the negative predictive
value was 98.7%. Hence their hot nodes can predict the nodal
status and the identification of their hot nodes as SNs is
feasible.
Figure 2. Distribution of absolute radioactivity counts.
DETECTION CRITERIA OF SNS BY ABSOLUTE RADIOACTIVITY
COUNT
For the 183 patients in whom a hot node was identified, the
mean ± SD absolute radioactivity counts were 655 ± 93 cps
(range, 10–8000 cps) for the hot spot ex vivo and 4.4 ± 0.9 cps
(range, 1–50 cps) over the basin after excision. The distribution of absolute radioactivity in 312 hot nodes and their
backgrounds is shown in Fig. 2. The counts showed a wide
distribution and the histograms do not show the possibility of
making a borderline distinction between the hot nodes and the
background for the detection of SNs.
DETECTION CRITERIA OF SNS BY THE COUNT RATIO
The count ratios of hot nodes and background were 283 ± 33
(range, 10–1800) and there was also a wide distribution (Fig.
3). When we defined an SN as having a count ratio of 100 or
more from the background level, because this ratio is sufficiently large for easy detection with the probe, only 169 hot
nodes (54.3%) met our definition of the SN. In other words,
SNs were not identified in 60 patients (45.5%) whose hottest
node count ratio was less than 100.
When SNs were defined as having a radioactivity count ratio
of 10 or more, all 312 hot nodes met our definition of SNs and
non-hot nodes remained as non-SNs.
DISCUSSION
Intraoperative lymphatic mapping was introduced by our institution as a means of selecting patients suitable for undergoing
axillary dissection using radiocolloids (6–8). The success in
identifying the SN is dependent on the large difference in
radioactivity between hot nodes and the axillary basin. However, this method is not objective and it is difficult to offer a
more accurate procedure without the use of more concrete SN
Figure 3. Distribution of hot node/background count ratio.
detection criteria. In this study, the SN detection criteria were
examined on the SNs in the axillary area because SNs around
the peritumoral area were not removed in our approach.
Both Loggie et al. (9) and Wong et al. (10) defined the SN in
patients with melanoma in terms of the level of radioactivity
alone. However, absolute counts are time dependent from the
moment of injection of the radiocolloid and are not easily
reproducible using a gamma probe, and using this criterion
may be misleading because the hottest node may vary with
patient age (11). In fact, according to our results, there was no
borderline distinction between the hot nodes and the background and it was impossible to define the SN in terms of
absolute radioactivity.
Large-particulate agents, such as tin colloids, are trapped in
the SN only and are retained for a sufficiently long period (8).
The radioactive count ratios may not drop with longer delays,
while their absolute radioactivity diminishes. Therefore, we
prefer to use the count ratio as our criterion for detecting the
SN.
Although we are not the first to establish SN detection criteria in terms of count ratio, other investigators have used count
ratios that are subject to variability, especially in melanoma
(9,10,12–18). Krag et al. (12) adopted SN detection criteria
involving an absolute count of at least 15 per 10 s and a count
406
Detection criteria for SN using RI
ratio of ³3. Mudun et al. (18) used an absolute count of 300–
3000 per 10 s and a count ratio of ³30. Albertini et al. (13) initially suggested SN detection criteria using an in vivo count
ratio of ³3 and an ex vivo SN:non-SN ratio of ³10. However,
more recently, the SN was redefined by an in vivo count ratio
of ³2 (19).
A few reports have referred to SN detection criteria in breast
cancer, and the criteria varied with each institution. Krag et al.
(4) reported a method based on absolute radioactivity in which
a hot spot was defined as an area of localized radioactivity
separate from the injection site with counts of at least 25 per
10 s; the measurement was performed before incision was made.
Cox et al. (20) used criteria involving an ex vivo radioactivity
count ratio of hot nodes to non-hot nodes of 10 or an in vivo
radioactivity count ratio of hot nodes to the background of 3.
Because the techniques of SN identification vary and the criteria also vary, it is crucial that each institution reviews the data
on SN mapping and clarifies its criteria for detecting the SN in
their technique. The collection and analysis of these outcomes
could result in the standardization of an SNB technique.
References
1. Morton DL, Wen D-R, Wong JH, Economou JS, Cagle LA, Sto FK, et al.
Technical details of intraoperative lymphatic mapping for early stage
melanoma. Arch Surg 1992;127:392–9.
2. Krag DN, Weaver DL, Alex JC, Fairbank JT. Surgical resection and
radiolocalization of the sentinel lymph node in breast cancer using a
gamma probe. Surg Oncol 1993;2:335–9.
3. Giuliano AE, Kirgan DM, Guenther JM, Morton DL. Lymphatic mapping and sentinel lymphadenectomy for breast cancer. Ann Surg
1994;220:391–401.
4. Krag D, Weaver D, Ashikaga T, Moffat F, Klimberg VS, Shriver C, et al.
The sentinel node in breast cancer: a multicenter validation study. N Engl
J Med 1998;339:941–6.
5. Lucci A, Kelemen PR, Miller C, Chardkoff L, Wilson L. National practice
patterns of sentinel lymph node dissection for breast carcinoma. J Am Coll
Surg 2001;192:453–8.
6. Sato K, Hiraide H, Uematsu M, Tamaki K, Ishikawa H, Yamasaki T, et al.
Efficacy and significance of sentinel lymph node identification with
technetium-99m-labeled tin colloids for breast cancer. Breast Cancer
1998;5:389–93.
7. Sato K, Uematsu M, Saito T, Ishikawa H, Yamasaki T, Tamaki K, et al.
Indications and technique of sentinel lymph node biopsy in breast cancer
using 99m-technetium labeled tin colloids. Breast Cancer 2000;7:95–8.
8. Sato K, Uematsu M, Saito T, Ishikawa H, Tamaki K, Tamai S, et al. Sentinel lymph node identification for patients with breast cancer using large
size radiotracer particles – technetium-99m labeled tin colloids produced
excellent results. Breast J 2001;7:388–91.
9. Loggie BW, Hosseinian AA, Watson NE. Prospective evaluation of selective lymph node biopsy for cutaneous malignant melanoma. Am Surg
1997;63:1051–6.
10. Wong JH, Terada K, Ko P, Coel MN. Lack of effect of particle size on the
identification of the sentinel node in cutaneous malignancies. Ann Surg
Oncol 1998;5:77–80.
11. Sato K, Tamaki K, Tsuda H, Kosuda S, Kusano S, Hiraide H, et al. Clinicopathologic and technical factors associated with uptake of radiocolloid
by sentinel nodes in patients with breast cancer. Submitted.
12. Krag DN, Meijer SJ, Weaver DL, Loggie BW, Harlow SP, Tanabe KK, et
al. Minimal-access surgery for staging of malignant melanoma. Arch Surg
1995;130:654–60.
13. Albertini JJ, Cruse CW, Rapaport D, Wells K, Ross M, DeConti R, et al.
Intraoperative radiolymphoscintigraphy improves sentinel lymph node
identification for patients with melanoma. Ann Surg 1996;223:217–24.
14. Pijpers R, Collet GJ, Meijer S, Hoekstra OS. The impact of dynamic
lymphoscintigraphy and gamma probe guidance on sentinel node biopsy
in melanoma. Eur J Nucl Med 1995;22:1238–41.
15. Taylor A Jr, Murray D, Herda S, Vansant J, Alazraki N. Dynamic lymphoscintigraphy to identify the sentinel and satellite nodes. Clin Nucl Med
1996;21:755–8.
16. Wong JH, Truelove K, Ko P, Coel MN. Localization and resection of an
in transit sentinel lymph node by use of lymphoscintigraphy, intraoperative lymphatic mapping and a hand-held gamma probe. Surgery
1996;120:114–6.
17. O’Brien CJ, Uren RF, Thompson JF, Howman-Giles RB, PetersenSchaefer K, Shaw HM, et al. Prediction of potential metastatic sites in
cutaneous head and neck melanoma using lymphoscintigraphy. Am J Surg
1995;170:461–6.
18. Mudun A, Murray DR, Herda SC, Eshima D, Shattuck LA, Vansant JP, et
al. Early stage melanoma: lymphoscintigraphy, reproducibility of sentinel
node dissection and effectiveness of the intraoperative gamma probe.
Radiology 1996;199:171–5.
19. Glass LF, Messina JL, Cruse W, Wells K, Rapaport D, Miliotes G, et al.
The use of intraoperative radiolymphoscintigraphy for sentinel node
biopsy in patients with malignant melanoma. Dermatol Surg 1996;
22:715–20.
20. Cox CE, Salud CJ, Cantor A, Bass SS, Peltz ES, Ebert MD, et al. Learning
curves for breast cancer sentinel lymph node mapping based on surgical
volume analysis. J Am Coll Surg 2001;193:593–600.