The effect of sentinel lymph node biopsy on the Nottingham Prognostic Index in breast cancer patients

G. CSERNI

B·cs-Kiskun County Teaching Hospital, Department of Pathology, H-6000 Kecskemét, Hungary

Introduction Methods Results

Discussion Acknowledgements References

The Nottingham Prognostic Index (NPI) is a prognostic variable suitable for the stratification of breast cancer patients for adjuvant systemic treatment. The impact of sentinel node biopsy (SNB) and of the assessment of nodal involvement was evaluated in 136 successful SNB procedures completed by axillary dissection (AD). The three strategies assessed included SNB and AD in all cases, AD only in SN-positive cases, and AD only if the SN contained macrometastases. Isolated tumour cells in lymph nodes were regarded either as metastases or as negative findings. The success rate and accuracy of SNB were 90% and 96%, respectively. The NPI was influenced by variations in the surgical staging strategy and the definition of nodal involvement, in at most, five patients. Adjuvant systemic treatment, indicated on the basis of the NPI is less influenced by staging strategies and definitions of metastases than that given on the basis of nodal status alone.

Key words: Adjuvant systemic treatment, breast cancer, Nottingham Prognostic Index, sentinel lymph nodes

J.R.Coll.Surg.Edinb., 46, August 2001, 208-212

INTRODUCTION

The Nottingham Prognostic Index (NPI) is a composite prognostic parameter involving both time-dependent factors and aspects of biological aggresiveness. It was derived from the retrospective statistical (multi-variate) analysis of a number of clinicopathological parameters in 387 breast cancer patients.¹ The NPI was later prospectively validated in 320 patients, from the same institution.² Its value in prognostication has been documented by its application to a larger population of 1662 patients with breast cancer.³ Two studies from different institutions have confirmed the value of the NPI. ^{4, 5}

Patients can be divided into three prognostic groups on the basis of the NPI: a good prognostic group (GPG; NPI • 3.4), a moderate prognostic group (MPG; 3.4 < NPI • 5.4), and a poor prognostic group (PPG; NPI > 5.4). An excellent prognostic group (EPG; NPI • 3) can be distinguished within the GPG, and the MPG is also sometimes divided into two subgroups. The many advantages that a prognostic index may offer include the important one that it allows for the stratification of patients for therapy.6 Patients in the GPG may not need adjuvant systemic treatment, while it is generally advised that those patients with oestrogen receptor (ER) -expressing tumours in the MPG and the PPG should receive adjuvant endocrine treatment. Those patients whose tumours lack ER expression and fall in the MPG may receive adjuvant chemotherapy, depending on other prognostic factors. Adjuvant chemotherapy is usually indicated in ER-negative tumours in the PPG (provided the patient is able to tolerate this treatment).

The NPI is calculated via the following formula: 0.2 x size (cm) + grade (1-3) + nodal stage (1-3). The grade is defined by the Nottingham modification of the Bloom and Richardson system ^7,8 The nodal stage is scored as follows: 1 -node-negative cases; 2 - involvement of 1-3 low axillary nodes or an internal mammary node (for medial tumours); 3: -involvement of >3 axillary nodes and/or the apical node, or any low axillary node and an internal mammary node together. The definition of nodal involvement and the surgical strategy used for staging may alter the assignment of a given patient to the adequate prognostic group and, therefore, may alter her systemic treatment. Many studies demonstrate that the sentinel lymph node (SLN) accurately predicts the overall axillary nodal status in breast cancer patients and that axillary dissection is not necessary in SLN-negative disease. ^9-13 A few studies suggest that the involvement of non-SLNs can be predicted in some cases on the basis of SLN metastasis size and primary tumour size; thus, minimal involvement of SLNs may allow the omission of axillary dissection. ^14-17

From a series of lymphatic mapping and SLN biopsies completed with axillary dissection, we have assessed the possible influence of the surgical strategy and the interpretation of the histopathological findings on the NPI.

MATERIALS AND METHODS

Lymphatic mapping was performed by means of the peritumoural injection of patent blue dye (Patentblau V2.5%, Byk Gulden, Konstanz, Germany), either alone or supplemented in the later phase of the study by the peritumoural and/or periareolar injection of 99mTc-labelled colloidal human albumin. ¹⁸ Three patients received the blue dye by the peri-areolar injection technique and the colloid by peri-tumoural injection. Radioactive nodes were detected by means of an intra-operative gamma probe (C-Trak, Carewise, Morgan Hill, California). All blue and/or hot nodes were identified as SLNs, and because this was our feasibility study, the biopsy was immediately followed by axillary dissection in all but one invasive carcinoma. SLNs were serially sectioned at 50-100 or 250-mm intervals and stained with haematoxylin and eosin (HE). All patients with negative findings on HE had their SLNs immunostained for epithelial markers (cytokeratin: NCL-PAN-CK, Novocastra, Newcastle, UK or MNF116, Dako, Glostrup, Denmark; and epithelial membrane antigen: MU-182-UC, BioGenex, San Ramon, USA, for some cases at the beginning of the study). Non-SLNs were processed conventionally.

The tumours were staged by using the American Joint Committee on Cancer (AJCC) and Union International Contre le Cancer (UICC) common TNM system. ¹⁹ Suggested improvements to the system were also used. ^{20, 21} The SLN involvement (independently of any further nodal involvement) is referred to as pN (sn). Minimal involvement of the lymph nodes that did not meet the minimum criteria of a micrometastasis was referred to as pN0(i+). ^{20, 22}

The following staging strategies were assessed for defining nodal status and subsequently calculating the NPI.

A: Any tumour cells regarded as positive; all nodes assessed.
B: Any tumour cells regarded as positive; SLNs and dissected axillary nodes considered, dissection performed only if the SLN is positive.
C: Any tumour cells regarded as positive; SLNs and dissected axillary nodes considered, dissection performed only if the SLN is positive and the risk of the involvement of non-SLNs is high, i.e. pN1b (sn).
D: Only metastases as defined by the TNM system ^{19, 20} regarded as positive (pN0(i+) regarded as negative); all nodes assessed.
E: Only metastases as defined by the TNM system ^{19, 20} regarded as positive (pN0(i+) regarded as negative); SLNs and dissected axillary nodes considered, dissection performed only if the SLN is positive.
F: Only metastases as defined by the TNM system ^{19, 20} regarded as positive (pN0(i+) regarded as negative); SLNs and dissected axillary nodes considered, dissection performed only if the SLN is positive and the risk of the involvement of non-SLNs is high, i.e. pN1b (sn).

Table 1: False-negative cases assessed by SLN procedure

* All cases with false-negative SLNs were mapped with the vital blue dye method, but FN3 and FN6 were also examined by lymphoscintigraphy. SLN: sentinel lymph node; CK: cytokeratin; EIC: extensive intra-ductal component.

RESULTS

Sentinel lymph node biopsy was attempted in 160 patients diagnosed as having breast cancer (two bilateral cases, and hence 162 procedures) between August 1997 and February 2000. At least one SLN was identified in 145 procedures (89.5% success rate). Most failures occurred during the learning phase of lymphatic mapping. ¹⁸ One SLN biopsy was not followed by axillary dissection. The SLNs accurately predicted the nodal status of the axilla in 138 cases (95.8% accuracy). There were 6 false-negative SLN biopsy procedures (Table 1).

Eight cases were deleted from the study as no invasive cancer was identified (in situ ductal carcinomas and fibroadenomas) in the breast specimen, or the microinvasive part was unsuitable for grading and consequently for determination of the NPI. The pT and pN categories of the remaining 136 patients are displayed in Table 2. The median tumour size was 2.2 cm. All cases were M0 at the time of biopsy.

The numbers of patients in the various NPI-based prognostic groups, as determined by the different strategies listed in the Methods, are given in Table 3.

DISCUSSION

Nodal involvement is the main indicator for adjuvant systemic treatment in breast cancer patients. However, it is now generally accepted that a proportion of the node-negative patients may also benefit from this form of therapy. Patients are currently selected for adjuvant systemic treatment on the basis of multiple parameters, including the nodal stage, menopausal status, and the size, steroid receptor content and histological grade of the primary tumour. The NPI furnishes an opportunity to categorise patients according to their expected prognosis and to plan their adjuvant treatment on the basis of the expected benefit versus the risks and costs of the therapy. It is well established that the grading of breast carcinomas involves a subjective component, but the procedure is reasonably reproducible. ^{23, 24} Despite this subjectivity, the NPI retains its prognostic power and has been validated in independent institutional studies. The present study illustrates the influence of the SLN biopsy and nodal staging strategy on the NPI.

Table 2: Distribution of the tumours, as assessed by the pTNM system

*The pN1b category is divided into cases with 1-3, and those with >3 involved nodes; this subclassification is somewhat different from that in the pTNM system, as the cases listed in these columns could include not only categories pN1bi and pN1bii of the pTNM system, respectively, but also cases with extra-capsular extension of tumour growth (pN1biii) and those with metastatic nodes larger than 2 cm (pN1biv). The NPI does not account for the two latter features, but only for the number of involved nodes. **Another microinvasive cancer unsuitable for grading had isolated tumor cells in the SLN (pN0, i+)

Sentinel lymph node biopsy in breast cancer patients has been claimed to offer better staging than the conventional histology of axillary dissection specimens. ^25-28 This statement is true when the detection of micrometastases is considered, however, the significance of micrometastases remains controversial. A recent review and a large multi-institutional study suggest that the presence of micrometastases indicates a worse prognosis than that suggested by the presence of tumour-free nodes in patients with breast cancer. ^{29, 30} However, the definition of micrometastasis is not always well defined; some studies may include isolated tumour cells (ITCs) within this category. ²⁰ The documentation of ITCs, revealed by an extensive search of the lymph nodes, and their assignment as true nodal metastases may lead to the false upstaging of these patients. While ITCs in lymph node sinuses may later progress to metastases within nodes, their ability to survive and/or disseminate is poorly defined. It has been suggested, therefore, that nodes with ITCs should be regarded as free of mestatic involvement. ^{20, 22}

The survival data for colorectal carcinoma patients with ITCs in the bone marrow indicate that such tumour cells are not equivalent to metastases, and should not lead to the categorisation of such patients as having stage IV disease. ²⁰
Although ITCs detected by immunohistochemistry in the bone marrow have been shown to indicate a higher risk of distant relapse and cancer-related death in breast carcinoma patients, in the short-term, they are not considered to be an indication for systemic chemotherapy because of their high incidence and the lack of long-term follow-up data. The significance of ITCs in axillary nodes of breast cancer patients also awaits further studies. Since the possibilities of SLN involvement by ITCs or by micrometastases are not distinguished in most studies on SLNs, they are regarded as a single entity in this discussion. ^31-33

It appears that some institutions include metastases detected by a thorough search in the SLNs in their strategy for adjuvant systemic treatment, while others are more cautious in this respect. ^34-36 The interpretation of ITCs and micrometastases is a matter of continuing debate and is reflected by approaches A to C versus D to F of this study (the latter do not view ITC involvement as nodal mestastes).

The surgical strategy may also affect the nodal staging and the NPI. Axillary clearance (once considered the gold standard of staging) coupled with an intensive work-up of the SLNs is probably the best histological staging procedure for lymph node involvement. Its sensitivity is high and its costs are more reasonable than those of intensive histopathology of all axillary nodes. This strategy of staging was considered first (A and D).

The main perspective of SLN biopsy is the omission of axillary dissection in SLN-negative patients; strategies B and E consider this perspective, leaving the 6 false-negative SLN biopsy cases in the NPI range defined by a nodal score of 1.

Some studies suggest that axillary dissection may be unnecessary even if the SLNs contain micrometastases, since the SLNs are the only nodes involved in most of these patients. ^14-17 While results on this issue are more controversial, this strategy, incorporating the errors of both false-negative cases and the predictive model of the lack of non-SLN metastases in cases involving SLN micrometastases, has also been considered (C and F).

Table 3: Numbers of patients in the prognostic groups of the NPI in association with different staging strategies (n = 136)

* Note: EPG is a subgroup of GPG. For abbreviations, see text

The greatest difference between two strategies differing only in their definition of metastasis (e.g. A and D) is four patients. This would have resulted in, at most, 3% more patients in the GPG not requiring systemic adjuvant treatment in the latter group (D). The greatest difference between the two strategies, differing only in the surgical staging of the axilliary nodes (A and C), is five patients; this is also the greatest difference between any compared groups (A to F). This would have resulted, in at most, 4% more patients in the GPG. Although these differences are not substantial, they indicate the need for clear definitions as to what should be considered a metastasis and what should not. The Nottingham group currently includes any detected tumour cell in a lymph node as nodal involvement, but they do not search extensively for metastases. (This is purely a pragmatic approach to a difficult problem of interpretation and is a consequence of the paucity of data on clinical significance of micrometastases, their definition and the methodology used for their identification) (Ellis and Elston, personal communication). Their protocol involves the HE investigation of up to four levels of all larger nodes, while the surgical procedure is often a four-node sampling of the axillary contents. None of our investigated strategies can be equated with this approach, but approach A may be closer in principle. A randomised approach would probably be the best way to decide how minute metastatic foci and differences in surgical staging procedures should influence decisions on adjuvant systemic treatment.

A further consideration must be made. SLN biopsy has a well-documented learning curve and with, experience, the identification rate increases and the false negative rate decreases. ^13,39 By omitting the first 40 cases (commonly accepted as a learning phase), only one false negative biopsy could have been deleted, and this would not have substantially influenced the reported results. Despite the fact that successful series of SLN biopsy with dye only methods have been reported, the combined method involving both a dye and gamma probe guidance gives the best rates of identification and, therefore, ¹³ is the currently recommended method for SLN biopsy. Our series included only 25 cases mapped with this dual labelling method; this is why no attempt was made to separate patients on the basis of the biopsy technique used. However, it can be hypothesised that the combined biopsy technique could further reduce the influences of SLN biopsy strategy and interpretation of pathological findings on the NPI.

The NPI is a strong prognostic marker, which is also appropriate for selecting patients for further treatment. It is somewhat influenced by the subjectivity of assessing the histological grade of breast carcinomas; despite this subjectivity its prognostic strength remains unaltered. The present study also demonstrates the influence on the NPI of definitions of what constitutes a metastasis and different surgical approaches involving SLN biopsy. These influences are minor (smaller than influences on nodal status alone) and we feel that they will not have a significant impact on the NPI as a prognostic marker and a determinant of further adjuvant treatment. Stated otherwise, the NPI is robust even when SLN biopsy is used. The administration of adjuvant systemic treatment on the basis of the NPI is less influenced by variations of the surgical staging strategy and definitions of nodal metastasis than the same treatment given solely on the basis of the SLN and nodal status.

ACKNOWLEDGEMENTS

The work was supported by a J·nos Bolyai Research Fellowship of the Hungarian Academy of Sciences. The ideas and technical help of Drs. I Bodi, M Rajt·r and G Boross are also acknowledged.

REFERENCES

1. Haybittle JL, et al. A prognostic index in primary breast cancer. Br J Cancer 1982; 45: 361-6
2. Todd JH, et al. Confirmation of a prognostic index in primary breast cancer. Br J Cancer 1987; 56: 489-92
3. Elston CW, Ellis IO, Goulding H, Pinder SE. Role of pathology in the prognosis and management of breast cancer. In: Elston CW, Ellis IO, eds. The Breast, Systemic Pathology, 3rd edn, Volume 13. Edinburgh: Churchill Livingstone, 1998: 385-434
4. Brown JM, Benson EA, Jones M. Confirmation of a long-term prognostic index in breast cancer. Breast 1993; 2: 144-7
5. Balslev I, Axelsson CK, Zedeler K, Rasmussen BB, Carstensen B, Mouridsen HT. The Nottingham Prognostic Index applied to 9,149 patients from the studies of the Danish Breast Cancer Cooperative Group (DBCG). Breast Cancer Res Treat 1994; 32: 281-90
6. Blamey RW. Clinical aspects of malignant breast lesions. In: Elston CW, Ellis IO, eds. The Breast, Systemic Pathology, 3rd edition, Volume 13. Edinburgh: Churchill Livingstone, 1998: 501-14
7. Elston CW, Ellis IO. Assessment of histological grade. In: Elston CW, Ellis IO, eds. The Breast, Systemic Pathology, 3rd edition, Volume 13. Edinburgh: Churchill Livingstone, 1998: 365-84
8. Henson DE, Ries L, Freedman LS, Carriaga M. Relationship among stage of disease and histological grade for 22,616 cases of breast cancer. Breast Cancer Res Treat 1991; 22: 207-19
9. Giuliano AE, Jones RC, Brennan M, Statman R. Sentinel lymphadenectomy in breast cancer. J Clin Oncol 1997; 15: 2345-50
10. Krag D, et al. The sentinel node in breast cancer. A multi-center validation study. N Engl J Med 1998; 339: 941-6
11. Bass SS, Cox CE, Ku NN, Berman C, Reintgen DS. The role of sentinel lymph node biopsy in breast cancer. J Am Coll Surg 1999; 189: 183-94
12. Veronesi U, et al. Sentinel lymph node biopsy and axillary dissection in breast cancer: results in a large series. J Natl Cancer Inst 1999; 91: 368-73
13. McIntosh SA, Purushotham AD. Lymphatic mapping and sentinel node biopsy in breast cancer. Br J Surg 1998; 85: 1347-56
14. Chu KU, Turner RR, Hansen NM, Brennan MB, Bilchik A, Giuliano AE. Do all patients with sentinel node metastasis from breast carcinoma need complete axillary node dissection? Ann Surg 1999; 229: 536-41
15. Chu KU, Turner RR, Hansen NM, Brennan MB, Giuliano AE. Sentinel node metastasis in patients with breast carcinoma accurately predicts immunohistochemically
16. Dauway EL, Giuliano R, Pendas S, Cantor A, Cox C, Reintgen D. Characteristics of the sentinel lymph node in breast cancer predict further involvement of higher echelon nodes in the axilla. Abstract. Eur J Nucl Med 1999; 26(Suppl.): S92
17. Reynolds C, et al. Sentinel lymph node biopsy with metastasis: can axillary dissection be avoided in some patients with breast cancer. J Clin Oncol 1999; 17: 1720-6
18. Cserni G, Boross G, Balt·s B. The value of axillary sentinel nodal status in breast cancer. World J Surg 2000; 24: 341-4
19. American Joint Committee on Cancer. Breast. In: Fleming ID, et al., eds. AJCC Cancer staging manual, 5th edn. Philadelphia: Lippincott-Raven, 1997: 171-80
20. Hermanek P, Hutter RVP, Sobin LH, Wittekind C. Classification of isolated tumor cells and micrometastasis. Cancer 1999; 86: 2668-73
21. Sobin LH. Frequently asked questions regarding the application of the TNM classification. Cancer 1999; 85: 1405-6
22. Page DL, Anderson TJ, Carter BA. Minimal solid tumor involvement of regional and distant sites. When is a metastasis not a metastasis? Cancer 1999; 86: 2589-92.
23. Frierson HF, et al. Interobserver reproducibility of the Nottingham modification of the Bloom and Richardson histological grading scheme for infiltrating ductal carcinoma. Am J Clin Pathol 1995; 105: 195-8
24. Robbins P, et al. Histological grading of breast carcinomas. A study of interobserver agreement. Hum Pathol 1995; 26: 873-9
25. Giuliano AE, Dale PS, Turner RR, Morton DL, Evans SW, Krasne DL. Improved axillary staging of breast cancer with sentinel lymphadenectomy. Ann Surg 1995; 180: 700-4
26. Jannink I, Fan M, Nagy S, Rayudu G, Dowlatshahi K.Serial sectioning of sentinel nodes in patients with breast cancer: a pilot study. Ann Surg Oncol 1998; 5: 310-4
27. Cserni G. Metastases in axillary sentinel lymph nodes in breast cancer as detected by intensive histopathological work up. J Clin Pathol 1999; 52: 922-4 28. Dowlatshahi K, Fan M, Bloom KJ, Spitz DJ, Patel S, Snider HC. Occult metastases in the sentinel lymph nodes of patients with early stage breast carcinoma. A preliminary study. Cancer 1999; 86: 990-6
29. Dowlatshahi K, Fan M, Snider HC, Habib FA. Lymph node micrometastases from breast carcinoma. Reviewing the dilemma. Cancer 1997; 80: 1188-97
30. Cote RJ, et al. Role of immunohistochemical detection of lymph-node metastases in management of breast cancer. Lancet 1999; 354: 896-900
31. Relihan N, McGreal G, Kelly J, Ryan D, O’Sullivan GC, Redmond HP. Combined sentinel lymph-node mapping and bone-marrow micrometastatic analysis for improved staging in breast cancer. Lancet 1999; 354: 129-30
32. Braun S, et al. Cytokeratin-positive cells in the bone marrow and survival of patients with stage I, II or III breast cancer. New Engl J Med 2000; 342: 525-33
33. Smith BL. Approaches to breast-cancer staging. New Engl J Med 2000; 342: 580-1
34. Ollila DW, Brennan MB, Giuliano AE. Therapeutic effect of sentinel lymphadenectomy in T1 breast cancer. Arch Surg 1998; 133: 647-51
35. McIntosh SA, Going JJ, Soukop M, Purushotham AD, Cooke TG. Therapeutic implications of the sentinel lymph node in breast cancer. Lancet 1999; 354: 570
36. Wall E van der. The sentinel node in breast cancer: implications for adjuvant treatment? Eur J Nucl Med 1999; 26(Suppl): S17-S19
37. Rahusen FD, Diest PJ van, Meijer S. Do all patients with sentinel node metastasis from breast carcinoma need complete axillary node dissection? Ann Surg 2000; 231: 615-6
38. Cserni G. Sentinel lymph node biopsy-based prediction of further breast cancer metastases in the axilla. Eur J Oncol (In press)
39. Cody HS 3rd, Hill ADK, Tran KN, Brennan MF, Borgen PI. Credentialing for breast lymphatic mapping: How many cases are enough? Ann Surg 1999; 229: 723-8 detectable nonsentinel node metastasis. Ann Surg Oncol 1999; 6: 756-61

Copyright date: 26th April 2001
Correspondence: G·bor Cserni, B·cs-Kiskun County Teaching Hospital, Department of Pathology, H-6000 Kecskemét, Hungary
E-mail: cserni@freemail.c3.hu

©2001 The Royal College of Surgeons of Edinburgh, J.R.Coll.Surg.Edinb.