Author:

Janet Fricker

Introduction

Sentinel lymph node biopsy (SLNB) in breast cancer is a diagnostic procedure that establishes whether cancer cells have spread from the primary breast tumour into the lymphatic system, and can indicate that surgery is unnecessary on lymph nodes in the armpit. It identifies the first lymph node into which lymphatic fluid drains from the tumour, using a radiotracer or blue dye, and limits the biopsy to that node, which is known as the sentinel.

The armpit contains 20‒25 lymph nodes and many were routinely removed in a procedure called axillary lymph node dissection when a mastectomy (breast removal) was carried out, or for staging the cancer to see if it had spread from the breast. Loss of so many lymph nodes often led to lymphoedema (swollen arms and hands from accumulating lymph fluid), pain, arm weakness and loss of movement and limitations of hand movement, which can be chronic and life lasting.

SLNB was trialled in the mid-1990s at the European Institute of Oncology in Milan by the Italian surgeon, Umberto Veronesi, a pioneer in breast conserving surgery, and Giovanni Paganelli, a specialist in nuclear medicine. It was motivated by Veronesi’s philosophy of using the ‘minimal effective’ rather than the ‘maximum tolerated’ approach to cancer treatment.

Proving the case

Halstead radical mastectomy – named after William Stewart Halsted, a US surgeon born in 1852 – was for many years the standard treatment for breast cancer; removal of underlying muscles and axillary lymph nodes were integral parts of the procedure. Most women had axillary spread at diagnosis of their breast cancer, partly because they were reluctant to come forward and there was little perceived benefit in early diagnosis.

Then, in the late 1960s, researchers started to question whether cancer spread only in a continuous way from the breast, and that it could instead be a systemic disease, as thinking about cancer biology and metastasis evolved. A pivotal trial by the National Surgical Adjuvant Breast and Bowel Project (NSABP) in the US, started in 1971 and led by surgeon and ‘cancer giant’ Bernard Fisher, found that patients who had only simple and not radical mastectomy, without axillary dissection, had no greater risk of distant disease or death than those undergoing the radical procedure (see also this 25 year follow-up report).

In turn, the 1980s and 1990s saw the introduction of breast conserving surgery with lumpectomy and radiation, and mammographic screening, with women being diagnosed much earlier with fewer lymph node metastases.

But extensive removal of axillary lymph nodes continued for diagnostic reasons, says Paganelli, as nodal involvement is a key element in the TNM (tumour, node, metastasis) staging system, used to guide treatment decisions for many cancers.

“Surgical removal of the lymph nodes had no intent to cure the patient; it was just used to gain an idea of whether there was metastasis in the lymph nodes, which made patients candidates for more aggressive treatment,” he says. “In about 70% of cases, especially when patients had tumours of less than 2 cm, lymph nodes were negative.”

Understanding of the anatomical layout of axillary lymph nodes was helped by Veronesi’s study of nodes removed from 1,446 women during breast cancer surgery, published in 1990. For the study, Veronesi asked his surgical team to place a different coloured stitch in lymph nodes taken from level I, II, and III of the axilla, allowing pathologists analysing the nodes to know exactly which level they originated from.

Results revealed that, out of 839 patients found to have lymph metastases, the first level was the site of metastases in 828, the second level in 364 and the third in 187. In only 11 cases were second and or third levels invaded without metastases at the first level.

“Umberto described lymph nodes in the axilla as being like a pearl necklace, with each node located one after the other,” explains Alberto Costa, a breast surgeon who had trained with Veronesi. “Cells originating from breast cancer reach the nearest node first and then move on progressively. They don’t skip nodes. So, if the lymph node located nearest to the cancer is healthy you can be confident that others are unaffected by cancer.” Veronesi’s finding is widely considered to have opened the way for using SNLB in breast cancer surgery.

Sentinel node: not just in breast cancer

The idea of a sentinel node as the gatekeeper to nodes further along the chain is usually attributed to Ramon Cabanas, a South America urologist now based in the US, in 1977 in work on carcinoma of the penis. This was pioneering work but as Robert Mansel, a British breast cancer surgeon, has pointed out, the first reported clinical study on sentinel node biopsy was published more than 20 years earlier in work in the US on the sentinel node in parotid tumours (see also this article on the history and validation of the sentinel node).

The introduction of blue dye mapping in melanoma by Donald Morton (1934–2014) at the John Wayne Cancer Institute, Santa Monica, US, was a key point in the general acceptance of SLNB. At the World Health Organisation’s Second International Conference on Melanoma in 1989, Morton described how he had injected patent blue or isosulfan blue intradermally at the site of the primary melanoma, then followed the channel by meticulous dissection, tracing the dye to the first draining node.

Morton went on to identify the sentinel node in 194 patients (out of 237) using isosulfan blue dye injected perioperatively at the melanoma sites. Non-sentinel nodes, he found, were the sole site of melanoma metastasis in just two patients.

The first use of blue dye mapping in breast cancer was performed in 1994 by Armando Giuliano, also then at the John Wayne Cancer Institute, who adapted injection volumes and time to incision, taking account of lymph drainage from the breast. In his case series of 174 patients, Giuliano was able to localise a sentinel node in 114 patients (66%), and in 109 of 114 patients who had sentinel nodes identified he showed positive correlation between the sentinel node and overall axillary node status.

From Giuliano’s work the concept that the sentinel node could be localised in breast cancer and could predict axillary lymph node status was established.

Trialling SLNB

Veronesi, who learned about SLNB after hearing Giuliano speak at a US meeting, was an early adopter as it aligned with his philosophy of maintaining the dignity and quality of life of patients. But at the European Institute of Oncology, Veronesi found that blue dye had a low success rate, with his team only able to locate the sentinel node in about 40% of patients. “The blue colour could be hard to visualise. We found that it was easy to make a mistake and contaminate the entire area,” says Costa.

To improve on this, Paganelli, who headed nuclear medicine at the European Institute of Oncology, was inspired to inject a radioactive tracer near the tumour and then use a handheld gamma ray probe to identify the sentinel node. “At first Veronesi was sceptical of my idea, but he said ‘I trust you, let’s see what happens,’” remembers Paganelli.

Initially, two surgeons performed SLNB with the technique, known as sentinel node lymphoscintigraphy. When they performed axillary clearance on 16 patients who had first undergone the sentinel node procedure, they found 100% concordance between sentinel node and entire axillary clearance results. “It quickly became evident that the radiotracer allowed surgeons to locate the sentinel node faster. In comparison with the blue dye technique, they didn’t have to cut so much tissue to find the node, which resulted in less scarring,” says Costa.

Next, Veronesi performed a pilot study with 163 women, published in 1997, and showed that the sentinel node identified by the radiotracer technique could accurately predict axillary lymph node status in 97.5% of cases.

In a second study comparing different radiotracers and injection modalities, the team at the European Institute of Oncology showed administration of large sized 99mTc-labeled colloid particles (200‒1000 nm) worked best and subdermal injection was more suitable than peritumoural.

Veronesi and colleagues undertook the first randomised trial comparing SLNB with total axillary dissection. In the single centre study in 1998 to 1999, 516 patients with breast cancers measuring less than 2 cm were randomised to either SLNB plus immediate axillary dissection or SLNB with no further axillary treatment, if the sentinel node was negative. Results after more than five years showed there was no difference between the groups in axillary recurrence, distant metastasis or survival, but that arm pain and mobility were far better in those undergoing only SLNB.

Ten year results supported the findings, reporting slightly better overall survival in the SNLB-only arm.

SNLB has became an integral part of conservative breast cancer treatment as it allows axillary dissection to be avoided in a large proportion of early breast cancer patients, while still providing prognostic information to guide adjuvant treatment. In 2014, the American Society of Clinical Oncology (ASCO) stated that most early breast cancer patients should have SNLB and that if the sentinel node was negative, axillary dissection should be avoided.

SLNB: the story continues

Questions then arose about situations where axillary lymph nodes were found to be positive. There were concerns that for patients with limited sentinel node involvement, axillary dissection might represent overly aggressive treatment. While sentinel node dissection accurately identified nodal metastasis in early breast cancer it remained unclear whether further nodal dissection had any influence on survival. Such uncertainty was greatest in cases where patients were prescribed systemic therapy that had the ability to treat occult cells wherever they were located.

The American College of Surgeons Oncology Group Z0011 trial, undertaken by Giuliano, randomised women with sentinel lymph node metastases identified by biopsy to undergo complete axillary node dissection or no further axillary treatment. Results showed no differences between the two groups in five year overall or disease-free survival, suggesting that, in this subgroup, axillary dissection may not be necessary after a positive sentinel node result. The authors concluded that the only additional information obtained from dissection was the number of nodes affected by metastases, but that this information was unlikely to change decisions about systemic therapy.

The IBCSG 23-01 trial, conducted by the International Breast Cancer Study Group, which recruited patients with tumours up to 5 cm, showed that axillary dissection conferred no advantage in disease free survival and overall survival, as long as only micro-metastases (foci up to 2 mm) were present in the sentinel node.

As a consequence of Z0011 and IBCSG 23-01, the 2013 St Gallen International Consensus Panel, convened every two years in St Gallen, Switzerland, endorsed omission of axillary dissection in patients with one to two involved sentinel nodes undergoing breast-conserving surgery and radiotherapy.

And if axillary dissection can be omitted in some patients with positive sentinel nodes, is there any utility in performing SLNB at all? The fact that postsurgical therapy in modern breast cancer management is determined more on the basis of biologic tumour characteristics than nodal involvement made information derived from SNLB largely irrelevant.

The SOUND trial, one of the last trials to be initiated by Veronesi, was started at European Institute of Oncology in 2012, to explore the futility of performing SLNB in low-risk breast cancer patients. The study is randomising patients with early breast cancers of up to 2 cm who are candidates for breast-conserving surgery with a negative preoperative assessment of the axilla (using ultrasound to identify significant metastases) to either SLNB or observation alone, using the endpoint of disease-free survival.

If positive, the SOUND trial, expected to report in 2022, may signal the end of SLNB for people with low-risk cancers.