www.thelancet.com/oncology Published online August 6, 2024 https://doi.org/10.1016/S1470-2045(24)00350-4 1

Articles

Lancet Oncol 2024

Published Online

August 6, 2024

https://doi.org/10.1016/

S1470-2045(24)00350-4

SeeOnline/Comment

https://doi.org/10.1016/

S1470-2045(24)00385-1

*Group members are listed in the

appendix

Department of Molecular

Medicine and Surgery

(Prof J de Boniface MD PhD,

M Appelgren OCN,

Prof J Frisell MD PhD),

Department of Medical

Epidemiology and Biostatistics

(R Szulkin PhD), Karolinska

Institutet, Stockholm, Sweden;

Department of Surgery, Breast

Center, Capio St Göran’s

Hospital, Stockholm, Sweden

(Prof J de Boniface,

M Appelgren); Cytel,

Stockholm, Sweden (R Szulkin);

Faculty of Medicine, Institute

of Clinical Sciences, Lund

University, Lund, Sweden

(S Alkner MD PhD,

Prof L Rydén MD PhD);

Department of Hematology,

Oncology and Radiation

Physics, Skåne University

Hospital Lund, Lund, Sweden

(S Alkner); Department of

Surgery, Vastmanland Hospital

Vasteras, Vasteras, Sweden

(Y Andersson MD PhD); Centre

for Clinical Research, Uppsala

University and Region

Vastmanland, Vastmanland

Hospital Vasteras, Vasteras,

Sweden (Y Andersson,

Prof L Bergkvist MD PhD); Breast

Center Karolinska, Karolinska

Comprehensive Cancer Center,

Karolinska University Hospital,

Stockholm, Sweden

(Prof J Frisell); Breast Surgery,

IRCCS Ospedale San Raffaele,

Milano, Italy (O D Gentilini MD);

Vita-Salute San Raffaele

Completion axillary lymph node dissection for the

identification of pN2–3 status as an indication for adjuvant

CDK4/6 inhibitor treatment: a post-hoc analysis of the

randomised, phase 3 SENOMAC trial

Jana de Boniface, Matilda Appelgren, Robert Szulkin, Sara Alkner, Yvette Andersson, Leif Bergkvist, Jan Frisell, Oreste Davide Gentilini,

Michalis Kontos, Thorsten Kühn, Dan Lundstedt, Birgitte Vrou Offersen, Roger Olofsson Bagge, Toralf Reimer, Malin Sund, Peer Christiansen,

Lisa Rydén, Tove Filtenborg Tvedskov, on behalf of the SENOMAC Trialists’ Group*

Summary

Background In luminal breast cancer, adjuvant CDK4/6 inhibitors (eg, abemaciclib) improve invasive disease-free

survival. In patients with T1–2, grade 1–2 tumours, and one or two sentinel lymph node metastases, completion

axillary lymph node dissection (cALND) is the only prognostic tool available that can reveal four or more nodal

metastases (pN2–3), which is the only indication for adjuvant abemaciclib in this setting. However, this technique can

lead to substantial arm morbidity in patients. We aimed to pragmatically describe the potential benefit and harm of

this strategy on the individual patient level in patients from the ongoing SENOMAC trial.

Methods In the randomised, phase 3, SENOMAC trial, patients aged 18 years or older, of any performance status, with

clinically node-negative T1–T3 breast cancer and one or two sentinel node macrometastases from 67 sites in five

European countries (Denmark, Germany, Greece, Italy, and Sweden) were randomly assigned (1:1), via permutated

block randomisation (random block size of 2 and 4) stratified by country, to either cALND or its omission (ie, they had

a sentinel lymph node biopsy only). The primary outcome is overall survival, which is yet to be reported. In this posthoc analysis, patients from the SENOMAC per-protocol population, with luminal oestrogen-receptor positive, HER2-

negative, T1–2, histological grade 1–2 breast cancer, with tumour size of 5 cm or smaller were selected to match the

characteristics of cohort 1 of the monarchE trial who would only have an indication for adjuvant abemaciclib if found

to have 4 or more nodal metastases. The primary study objective was to determine the number of patients who

developed patient-reported severe or very severe impairment of physical arm function after cALND (as measured by

the Lymphedema Functioning, Disability, and Health [Lymph-ICF] Questionnaire) 1 year after surgery to avoid one

invasive disease-free survival event at 5 years with 2 years of adjuvant abemaciclib, using invasive disease-free survival

event data from cohort 1 of the monarchE trial. The SENOMAC trial is registered with ClincialTrials.gov, NCT02240472,

and is closed to accrual and ongoing.

Findings Between Jan 31, 2015, and Dec 31, 2021, 2766 patients were enrolled in SENOMAC and randomly assigned

to cALND (n=1384) or sentinel node biopsy only (n=1382), of whom 2540 were included in the per-protocol

population. 1705 (67%) of 2540 patients met this post-hoc study’s eligibility criteria, of whom 802 (47%) had a

cALND and 903 (53%) had a sentinel lymph node biopsy only. Median age at randomisation was 62 years

(IQR 52–71), 1699 (>99%) of 1705 patients were female, and six (<1%) were male. Among 1342 patients who

responded to questionnaires, after a median follow-up of 45∙2 months (IQR 25∙6–59∙8; data cutoff Nov 17, 2023),

patient-reported severe or very severe impairment of physical arm function was reported in 84 (13%) of 634 patients

who had cALND versus 30 (4%) of 708 who had sentinel lymph node biopsy only (χ² test p<0∙0001). To avoid one

invasive disease-free survival event at 5 years with adjuvant abemaciclib, cALND would need to be performed in

104 patients, and would result in nine patients having severe or very severe impairment of physical arm function

1 year after surgery.

Interpretation As a method to potentially identify an indication for abemaciclib, and subsequently avoid invasive

disease-free survival events at 5 years with 2 years of adjuvant abemaciclib, cALND carries a substantial risk of severe

or very severe arm morbidity and so cALND should be discouraged for this purpose.

Funding Swedish Research Council, the Swedish Cancer Society, the Nordic Cancer Union, and the Swedish Breast

Cancer Association.

Copyright © 2024 Elsevier Ltd. All rights reserved, including those for text and data mining, AI training, and similar

technologies.

Articles

2 www.thelancet.com/oncology Published online August 6, 2024 https://doi.org/10.1016/S1470-2045(24)00350-4

University, Milano, Italy

(O D Gentilini); 1st Department

of Surgery, National and

Kapodistrian University of

Athens, Athens, Greece

(M Kontos MD); Die Filderklinik,

Breast Center, Filderstadt,

Germany (Prof T Kühn MD PhD);

Department of Gynecology and

Obstetrics, University of Ulm,

Ulm, Germany (Prof T Kühn);

Department of Oncology,

Institute of Clinical Sciences,

Sahlgrenska Academy at

Gothenburg University,

Gothenburg, Sweden

(D Lundstedt MD PhD);

Department of Oncology

(D Lundstedt), Department of

Surgery

(Prof R Olofsson Bagge MD PhD),

Sahlgrenska University

Hospital, Gothenburg, Sweden;

Department of Oncology,

Aarhus University Hospital,

Aarhus University, Aarhus,

Denmark

(Prof B V Offersen MD PhD);

Department of Experimental

Clinical Oncology, Danish

Center for Particle Therapy,

Aarhus, Denmark

(Prof B V Offersen); Department

of Surgery, Institute of Clinical

Sciences, Sahlgrenska Academy

at Gothenburg University,

Gothenburg, Sweden

(Prof R Olofsson Bagge);

Department of Obstetrics and

Gynecology, University of

Rostock, Rostock, Germany

(Prof T Reimer MD PhD);

Department of Surgery,

University of Helsinki and

Helsinki University Hospital,

Helsinki, Finland

(Prof M Sund MD PhD);

Department of Diagnostics and

Intervention-Surgery, Umeå

University, Umeå, Sweden

(Prof M Sund); Department of

Plastic and Breast Surgery,

Aarhus University Hospital,

Aarhus, Denmark

(Prof P Christiansen MD DMSc);

Department of Clinical

Medicine, Aarhus University,

Aarhus, Denmark

(Prof P Christiansen);

Department of

Gastroenterology and Surgery,

Skåne University Hospital

Lund, Malmö, Sweden

(Prof L Rydén); Faculty of Health

and Medical Sciences,

University of Copenhagen,

Copenhagen, Denmark

(Prof T Filtenborg Tvedskov MD

PhD); Department of Breast

Surgery, Gentofte Hospital,

Introduction

Nodal status is a strong prognostic factor in breast cancer.

Before the clinical implementation of sentinel lymph

node biopsy, pathological nodal status in clinically nodenegative patients was determined by axillary lymph node

dissection (ALND), with substantial consequences for

arm morbidity and quality of life.1

ALND provides

detailed information on the number of affected nodes,

whereas sentinel lymph node biopsy offers less complete

nodal staging. Multiple attempts have been made to deescalate surgical staging of the axilla, and since the

implementation of sentinel lymph node biopsy the

number of clinical indications for a completion ALND

(cALND; ie, an ALND after sentinel node biopsy) has

been substantially reduced.1–4 This reduction in clinical

indications for cALND is especially relevant in this era,

when survival in breast cancer has increased substantially,

and more focus is being given to aspects of survivorship,

such as quality of life. The potential benefits of a cancer

treatment should thus be weighed against its functional,

physical, and psychosocial consequences.

In patients with clinically node-negative breast cancer

who have one or two sentinel lymph node metastases,

cALND does not improve survival.1–4 This is despite the

fact that patients with four or more nodal metastases

(pN2–3 status; 13∙7% and 12∙9% after cALND in

ACOSOG Z0011 and AMAROS, respectively), who might

have an indication for intensified adjuvant treatment

strategies, remain unidentified if they have a sentinel

lymph node biopsy only. Importantly, arm lymphoedema

and patient-reported arm swelling is significantly less

common after axillary radiotherapy than after cALND.1,5

In the monarchE6,7 and NATALEE8

trials, the addition of

adjuvant CDK4/6 inhibitors to standard endocrine

therapy resulted in improved invasive disease-free

survival. The recently published NATALEE trial8

included

patients with both node-positive and node-negative

luminal HER2 (also known as ERBB2)-negative breast

cancer to evaluate survival benefits through the addition

of adjuvant ribociclib to endocrine treatment, the

monarchE trial, which tested the CDK4/6 inhibitor

abemaciclib, limited enrolment to only high-risk

patients.9

High risk in the monarchE trial was defined as

either pathological N stage 2–3 (pN2–3) or one-to-three

nodal metastases (pN1) in combination with additional

risk factors—namely, histological grade 3 or tumour size

larger than 5 cm. Hence, in patients with luminal

HER2-negative breast cancer undergoing primary

surgery that identified one-to-three nodal metastases, but

without additional risk factors, a cALND might need to

Research in context

Evidence before this study

We searched PubMed on Oct 13, 2023, and again on

March 30, 2024, for English-language studies published since

database inception, using the terms “abemaciclib”, “CDK4/6

inhibitor”, “randomized trial”, “axillary surgery”, and “axillary

(lymph node) dissection”, presenting trial results for the

addition of CDK4/6 inhibitors to adjuvant endocrine treatment

in luminal HER2-negative breast cancer, or discussing the

potential effect of such trial results on axillary surgery practice.

Studies regarding the use of CDK4/6 inhibitors in the

neoadjuvant or metastatic setting were disregarded. We

identified multiple publications of the monarchE trial and, more

recently, the NATALEE trial. We deduced that only the monarchE

trial eligibility criteria created a clinical dilemma by

necessitating a completion axillary lymph node dissection

(cALND) in patients with grade 1–2 breast tumours that were

smaller than 5 cm in diameter and one or two sentinel lymph

node metastases, such that they would only have an indication

for adjuvant abemaciclib in the case of identification of four or

more axillary nodal metastases. We identified only two

commentaries on the subject, discussing the value and

potential consequences of such an increased use of cALND after

a positive sentinel lymph node biopsy, but no studies reporting

on the effect of such practices for the patients. Given the

increasing evidence that cALND does not provide a survival

benefit for patients with one or two metastases in their sentinel

lymph node biopsy, and the resulting efforts of clinicians to

de-escalate axillary surgery and thus reduce patients’

impairment of arm function and the incidence of arm

lymphoedema, we aimed to pragmatically describe the

potential benefits and harms that patients would be exposed to

if receiving a cALND with the aim of identifying the few

individuals with four or more nodal metastases.

Added value of this study

By use of data from the per-protocol population of the large,

international, randomised, non-inferiority SENOMAC trial, we

identified a significantly increased occurrence of patientreported and objectively measured arm morbidity among

patients exposed to cALND, and a small change of improving

oncological outcomes with treatment with adjuvant

abemaciclib. By presenting numbers needed to treat, diagnose,

and harm, we provide easily applicable and pragmatic data that

can be used in the discussion with patients who need to

understand their risks and potential benefits when confronted

with the option of more extensive axillary surgery.

Implications of all the available evidence

The present analysis provides unique, patient-focused, clinically

relevant data that weigh survival outcomes from the monarchE

trial against arm morbidity data from the SENOMAC and

AMAROS trials. We propose that the high number of patients

potentially developing severe arm morbidity after a cALND,

among whom very few will have any oncological benefit as a

result, clearly disqualifies this staging tool for the identification

of candidates for adjuvant abemaciclib.

Articles

www.thelancet.com/oncology Published online August 6, 2024 https://doi.org/10.1016/S1470-2045(24)00350-4 3

Gentofte, Denmark

(Prof T Filtenborg Tvedskov)

Correspondence to:

Prof Jana de Boniface, Breast

Center, Capio St Goran’s Hospital,

11219 Stockholm, Sweden

jana.de-boniface@ki.se

See Online for appendix

be performed to identify candidates for adjuvant

abemaciclib. The consequences of such an escalation of

axillary surgery have not been evaluated from a

perspective of clinically relevant arm morbidity.10,11

The international SENOMAC trial randomly assigned

patients with clinically node-negative breast cancer and

one or two sentinel lymph node macrometastases to

cALND or its omission (ie, sentinel lymph node biopsy

only). The primary endpoint of overall survival has not

yet been reported, but non-inferiority regarding

recurrence-free survival has been published, showing

non-inferiority of the omission of cALND,12 and 1-year

patient-reported outcome measures showed significantly

worse arm morbidity after cALND compared with

sentinel lymph node biopsy alone.13 Here, we aimed to

weigh the numbers needed to diagnose by cALND to

identify individuals with pN2–3 against (1) the number

needed to treat for avoiding one invasive disease-free

survival event at 5 years after completing 2 years of

adjuvant abemaciclib and (2) the number needed to

harm regarding severe or very severe patient-reported

arm morbidity at 1 year and arm lymphoedema at

5 years.

Methods

Study design and participants

The randomised, phase 3 SENOMAC trial included

patients aged 18 years or older of any performance status,

both female and male (as defined by personal identity

numbers), with clinically node-negative T1–T3 breast

cancer and one or two sentinel node macrometastases

from 67 sites in five European countries (Denmark,

Germany, Greece, Italy, and Sweden). All patients

provided informed written consent. The trial was

registered at ClinicalTrials.gov, NCT02240472, and the

protocol has been published elsewhere.14 The trial was

approved by the Swedish Ethical Review Authority

(2014/1165-31/1) and relevant Ethical Review Boards in

participating countries and was performed and

monitored according to Good Clinical Practice.

For this post-hoc analysis, patients aged 18 years and

older undergoing primary surgery for luminal, oestrogen

receptor-positive, HER2-negative, T1–T2 breast cancer

(ie, ≤5 cm in diameter), and histological grades 1 or 2

were selected from the per-protocol population of the

SENOMAC trial, to match the eligibility criteria for the

cohort 1 of the monarchE trial, who would only have an

indication for adjuvant abemaciclib if found to have four

or more nodal metastases.6 The per-protocol population

included patients who had not withdrawn their informed

consent within 21 days from randomisation, had received

the axillary surgery they were randomly assigned to, and

had not violated eligibility criteria at enrolment. A preoperative axillary ultrasound was mandatory. Exclusion

criteria for SENOMAC were previous invasive breast

cancer, regional or distant metastases, bilateral breast

cancer if one side met exclusion criteria, medical

contraindications against radiotherapy or systemic

treatment, or an inability to understand the study

information. The present analysis was not specified in

the SENOMAC trial protocol but designed post-hoc and

was not affected by any protocol amendments.

Procedures

Full details of the SENOMAC procedures have been

published elsewhere.12,13 Briefly, all provisionally eligible

patients underwent sentinel lymph node biopsy showing

one or two macrometastases, and were then randomly

assigned (1:1) to cALND or its omission (ie, they had a

sentinel lymph node biopsy only) using permutated

block randomisation (random block size of 2 and 4),

stratified by country.

Adjuvant systemic therapy, including adjuvant

radiotherapy, was given in accordance with national

guidelines.

All patients completed questionnaires regarding

health-related quality of life and long-term complications

to the arm (the EORTC Quality of Life Questionnaire

[QLQ-C30], the breast cancer-specific QLQ-BR23

questionnaire, the EuroQol Group 5-Dimension

questionnaire, and the Lymphedema Functioning,

Disability, and Health Questionnaire [Lymph-ICF

questionnaire15]) at enrolment and will continue to

complete them at 1, 3, 5, and 10 years after surgery. No

clinical measures of lymphoedema were done.

For the present analysis of arm morbidity, 1-year

patient-reported outcome measures using the LymphICF questionnaire were extracted from the SENOMAC

database. The Lymph-ICF questionnaire, specifically

addressing impairment of arm function and activity

limitations, includes 29 questions producing a total score

and five separate domains (physical function, mental

function, household activities, mobility activities, and

social activities). Each question is answered on a visual

analogue scale and renders scores from 0 to 100, where

higher numbers indicate worse arm morbidity.16 Domain

scores are calculated as the average of all non-missing

individual items that belong to the relevant domain, and

the total score takes all items into account. No central

review of domain scores was done.

Outcomes

In the SENOMAC study, the primary endpoint is overall

survival. For the present post-hoc analysis, the primary

objective was the number of patients developing patientreported severe or very severe impairment of physical

arm function after cALND as measured by the LymphICF questionnaire 1 year after surgery to avoid one

invasive disease-free survival event at 5 years with 2 years

of adjuvant abemaciclib using data on invasive diseasefree survival from cohort 1 of the monarchE trial. To

obtain this number, the numbers needed to treat,

numbers needed to diagnose, and numbers needed to

harm were calculated separately.

Articles

4 www.thelancet.com/oncology Published online August 6, 2024 https://doi.org/10.1016/S1470-2045(24)00350-4

Secondary objectives were the expected rate of clinical

arm lymphoedema at 5 years and treatment for arm

lymphoedema during 5 years postoperatively using

published data from the AMAROS trial,1

which randomly

assigned patients with breast cancer and one or two

sentinel lymph node metastases to cALND or axillary

radiotherapy.

Statistical analysis

The number needed to treat estimates how many

individuals would need to receive a treatment to avoid

one negative event and is the inverse of the absolute risk

reduction.17,18 The absolute risk reduction is the control

event rate minus the experimental event rate, and thus

number needed to treat is equivalent to 1 divided by the

absolute risk reduction. Number needed to treat is

rounded to the nearest whole number. Relevant figures

were extracted from the recently published 5-year results

of the monarchE trial on invasive disease-free survival

(ie, an event was any death, ipsilateral or contralateral

invasive breast cancer recurrence, or secondary invasive

cancer). Only cohort 1 of monarchE was considered

(n=5120), which included patients with either four or

more nodal metastases or one-to-three nodal metastases

in combination with either grade 3 or a tumour of larger

than 5 cm in diameter, who were randomly assigned (1:1)

to adjuvant abemaciclib and endocrine treatment or to

endocrine treatment alone.7

We chose to only use cohort 1

of the monarchE trial because cohort 2 was smaller

(n=517) and was based on centrally determined Ki67 of

20% or more and the SENOMAC trial did not integrate a

central review of Ki67 and so could not allow for adequate

matching. Additionally, the US Food and Drug

Administration (FDA) approval of abemaciclib, which

was initially restricted to patients with a Ki67 of more

than 20% in 2021, was updated in March, 2023, to remove

the Ki67 testing requirement, and most guidelines do not

take Ki67 into account when defining indications for

adjuvant abemaciclib.

The number needed to diagnose calculates the number

of patients who need to undergo a diagnostic or staging

procedure—in this case a cALND—to identify one

additional individual with the state of interest—in this

case four or more nodal metastases. Here we calculated

number needed to diagnose as 1 divided by the difference

between the proportion of patients with four or more

nodal metastases in the cALND group and the

corresponding proportion in the sentinel lymph node

biopsy alone group.

The number needed to harm estimates how many

individuals would need to be exposed to a potentially

harmful treatment until one additional negative event is

observed, and is the inverse of the absolute risk increase.19

Like the absolute risk reduction, the absolute risk increase

is the control event rate minus the experimental event

rate, only that events are specified to be negative events

(ie, deaths and adverse events). Number needed to harm

is rounded to the nearest whole number. Number needed

to harm was calculated via two different strategies: (1) as

the number of eligible patients in SENOMAC who needed

to undergo a cALND to get one negative event, defined as

patient-reported severe or very severe impairment of

physical arm function, which was chosen since it most

closely resembles symptoms of arm lymphoedema, and

(2) as the number of patients from AMAROS who needed

to undergo a cALND to get one negative event, defined as

either clinical signs of lymphoedema at 5 years or

treatment for arm lymphoedema at any time during

5 years of follow-up. Updated AMAROS follow-up data on

5-year arm morbidity were published in 2022 (frequencies

on lymphoedema found in the data supplement of the

AMAROS publication).1

The primary endpoint of the SENOMAC trial itself,

overall survival, has not yet been reported. The hypothesis

is that overall survival will not be worsened by more than

2∙5% after 5 years, corresponding to a hazard ratio not

exceeding 1∙44 when comparing sentinel lymph node

biopsy alone with cALND. For the trial to have 80%

power with a one-sided α of 10%, a total of 190 deaths

need to occur with a target accrual of 3000 patients.

12

Figure: Inclusion of participants from the SENOMAC trial in the current analysis

*monarchE criteria were oestrogen-receptor positive, HER2-negative, tumour size of ≤5 cm, and histological

grade 1–2. In addition to the selected original SENOMAC subpopulation, 5-year arm lymphoedema rates were based

on the reported clinical sign of lymphoedema in 101 (24·5%) participants in the cALND group and in 45 (11·9%) in

the axillary radiotherapy group of the AMAROS trial.1

cALND=completion axillary lymph node dissection.

2766 participants randomly assigned in SENOMAC

403 did not fulfill monarchE criteria* 432 did not fulfill monarchE criteria*

1205 allocated to cALND in per-protocol

population

1335 allocated to sentinel lymph node

biopsy alone in the per-protocol

population

1384 assigned to cALND 1382 assigned to sentinel node biopsy

alone

802 included in current analysis

(number needed to diagnose)

634 questionnaire responders

included in current analysis

(number needed to harm)

903 included in current analysis

(number needed to diagnose)

708 questionnaire responders

included in current analysis

(number needed to harm)

179 excluded from per-protocol

population

131 withdrew consent within

21 days of random

assignment

36 did not meet eligibility

criteria

7 excluded within 21 days of

random assignment

5 did not receive assigned

treatment

47 excluded from per-protocol

population

11 withdrew consent within

21 days of random

assignment

32 did not meet eligibility

criteria

3 excluded within 21 days of

random assignment

1 did not receive assigned

treatment

Articles

www.thelancet.com/oncology Published online August 6, 2024 https://doi.org/10.1016/S1470-2045(24)00350-4 5

Descriptive variables are presented as frequencies and

percentages in the case of categorical variables, and as

mean (SD) or median (IQR) for continuous variables.

Lymph-ICF domain scores were categorised according to

the severity of arm morbidity into no or a small problem

(score of 0 to <25), a moderate problem (score of

25 to <50), and a severe or very severe problem (score of

50 to 100). Results are presented as numbers and

proportions among responders in each category, and

mean (SD) score per domain. The distribution of

categories over the randomisation groups was tested

with the two-sided χ² test. Mean values were compared

using Student’s t test. The questionnaire response rate

was calculated by dividing the number of completed

questionnaires by the number of individuals who had the

opportunity to complete the questionnaire. Patients who

had opted out of completing questionnaires due to

language difficulties or other reasons, or those treated at

sites that did not distribute questionnaires in the trial,

were not counted as potential responders. The

significance level was adjusted for multiple testing for

Lymph-ICF results using a Bonferroni correction, such

that p values equal to 0∙0087 or less were considered to

be significant.

All statistical analyses were performed using R

(version 4.1.2).

Role of the funding source

The funders of the study had no role in study design,

data collection, data analysis, data interpretation, or

writing of the report.

Results

Between Jan 31, 2015, and Dec 31, 2021, 2766 patients were

enrolled in SENOMAC and were randomly assigned to

cALND (n=1384) or sentinel node biopsy only (n=1382), of

whom 2540 were included in the SENOMAC per-protocol

population. For this analysis, we included 1705 (67%) of

2540 patients of the SENOMAC per-protocol population

who had received primary surgery, of whom 903 (53%) had

been assigned to sentinel lymph node biopsy alone and

802 (47%) had been assigned to cALND (figure). As of data

cutoff (Nov 17, 2023), median follow-up was 45·2 months

(IQR 25∙6–59∙8). Median age at randomisation was

62 years (IQR 52–71), 1699 (>99%) of 1705 patients were

female, and six (<1%) were male. Baseline clinical and

disease characteristics are shown in table 1. Race, ethnicity,

and self-reported gender were not registered.

The questionnaire response rate for the Lymph-ICF

questionnaire at 1 year of follow-up was 82∙4% with 1342

of 1629 potential responders, 634 (83%) of 764 in the

cALND group and 708 (82%) of 865 in the sentinel lymph

node biopsy only group. The domains of physical

function, mental function, mobility activities, and the

total score showed significant differences between the

groups on both statistical tests, with the sentinel lymph

node biopsy group having significantly better outcomes

Completion axillary

lymph node dissection

group (N=802)

Sentinel lymph node

biopsy alone group

(N=903)

Age, years

<65 451 (56%) 529 (59%)

≥65 351 (44%) 374 (41%)

Median 62 (53–70) 62 (52–71)

Sex

Female 800 (>99%) 899 (>99%)

Male 2 (<1%) 4 (<1%)

Tumour size, mm 20·3 (9·8) 21·5 (10·0)

Tumour stage

T1 (≤20 mm) 493 (61%) 521 (58%)

T2 (21–50 mm) 309 (39%) 382 (42%)

Sentinel node macrometastases

1 674 (84%) 770 (85%)

2 128 (16%) 133 (15%)

Number of lymph nodes removed 15·5 (7·0) 2·3 (1·5)

Number of axillary metastases 2·2 (2·4) 1·3 (0·5)

Breast surgery

Breast-conserving surgery 564 (70%) 612 (68%)

Mastectomy 238 (30%) 291 (32%)

Tumour histological type

Invasive carcinoma of no specific type 621 (77%) 669 (74%)

Lobular carcinoma 157 (20%) 200 (22%)

Other 24 (3%) 34 (4%)

Nottingham histological grade

Grade 1 199 (25%) 233 (26%)

Grade 2 603 (75%) 670 (74%)

Lymphovascular invasion

Yes 190 (24%) 215 (24%)

No 609 (76%) 676 (75%)

Missing 3 (<1%) 12 (1%)

Ki67 score

Mean 18·7 (12·1) 19·2 (12·5)

Missing 11 (1%) 10 (1%)

Adjuvant chemotherapy*

Yes 447 (56%) 470 (52%)

No 349 (44%) 428 (47%)

Missing 1 (<1%) 0

Adjuvant endocrine therapy*

Yes 785 (98%) 878 (97%)

No† 11 (1%) 20 (2%)

Missing 1 (<1%) 0

Radiotherapy*

None 41 (5%) 36 (4%)

Breast or chest wall only 51 (6%) 60 (7%)

Breast or chest wall plus regional lymph nodes 698 (87%) 789 (87%)

Regional lymph nodes only 5 (1%) 11 (1%)

Missing 2 (<1%) 2 (<1%)

Data are n (%), mean (SD), or median (IQR). Percentages might add up to more than 100% due to rounding.

*Ten patients (five in each group) terminating trial participation before their 1-year follow-up visit, at which adjuvant

treatment was reported, are excluded. †31 patients did not receive endocrine treatment due to patient wish (n=13),

depression (n=1), severe body pain (n=1), and unregistered reasons (n=16).

Table 1: Clinical characteristics of SENOMAC patients included in current analysis (N=1705)

Articles

6 www.thelancet.com/oncology Published online August 6, 2024 https://doi.org/10.1016/S1470-2045(24)00350-4

than the cALND group; however, there was no difference

for the domains of social activities and household

activities (table 2). Severe or very severe impairment of

physical arm function was significantly more common

after cALND (84 [13%] of 634) than after sentinel lymph

node biopsy alone (30 [4%] of 708; p<0∙0001), with an

absolute risk increase of 9%. The number needed to

harm was thus 11 (ie, 1 divided by 0∙09); such that

11 patients undergoing cALND resulted in one patient

with severe or very severe impairment of physical

arm function.

Five (1%) of 903 patients in the sentinel lymph node

biopsy alone and 101 (13%) of 802 patients in the cALND

group had four or more nodal metastases, an absolute

difference of 12%. In the sentinel lymph node biopsy

alone group, four or more nodal metastases were

detected on unintentional removal of additional nonsentinel lymph nodes without a regular cALND. The

number needed to diagnose was 8 (ie, 1 divided by 0∙12);

such that eight patients needed to undergo a cALND to

identify one candidate for adjuvant abemaciclib in

accordance with monarchE criteria. The published 5-year

invasive disease-free survival rates from cohort 1 in the

monarchE trial are 83∙2% (95% CI 81·5–84·7) in the

abemaciclib plus endocrine therapy group and 75∙3%

(95% CI 73·4–77·2) in the endocrine therapy alone

group, resulting in an absolute risk reduction of 7∙9%.6,7

The number needed to treat is thus 13 (1 divided by

0∙079); 13 patients need to be treated with abemaciclib to

avoid one invasive disease-free survival event at 5 years.

This finding implies that 104 (8 multiplied by 13) patients

would need to undergo a cALND to avoid one invasive

disease-free survival event at 5 years.

The AMAROS trial reported clinical signs of

lymphoedema after 5 years in 24∙5% in the cALND

group versus 11∙9% in the axillary radiotherapy group

(absolute risk increase of 12∙6%). The corresponding

figures for treatment for arm lymphoedema at any

timepoint up to 5 years were 36∙0% versus 20∙2%

(absolute risk increase of 15∙8%). Thus, the number

needed to harm for clinical signs of lymphoedema was

eight and for treatment of lymphoedema was 6 (1 divided

by 0∙126 and 0∙158, respectively).

This finding implies that if using cALND as a staging

tool, nine patients (104 divided by 11) would develop

severe or very severe impairment of physical arm

function, 13 patients (104 divided by 8) would develop

clinical signs of lymphoedema, and 17 patients

(104 divided by 6) would need lymphoedema treatment,

to avoid one invasive disease-free survival event at 5 years

if completing 2 years of adjuvant abemaciclib.

Discussion

In this analysis, we evaluated the potential consequences

of cALND in patients whose indication for adjuvant

abemaciclib solely depends on the diagnosis of pN2–3

status. We determined that 104 patients would need a

cALND to avoid one invasive disease-free survival event

at 5 years by receiving 2 years of adjuvant abemaciclib

and that nine patients would develop severe or very

severe impairment of physical arm function to avoid one

invasive disease-free survival event at 5 years. Although

adjuvant abemaciclib has become standard of care in

high-risk luminal breast cancer,7,9,20 it has not yet shown

an overall survival benefit.6,7 And so, in light of these

findings, which are in line with previous contributions to

this clinical debate,10,11,21 we discourage the use of cALND

in this setting.

Completion axillary

lymph node dissection

group (n=634)

Sentinel lymph node

biopsy alone group

(n=708)

p value

Physical function

Mean score 21·9 (21·6) 12·1 (16·1) <0·0001

No or a small problem 416 (66%) 599 (985%) <0·0001

Moderate problem 132 (21%) 72 (10%) ··

Severe or very severe problem 84 (13%) 30 (4%) ··

Missing 2 (<1%) 7 (1%) ··

Mental function

Mean score 13·9 (21·0) 8·3 (16·5) <0·0001

No or a small problem 503 (79%) 634 (90%) <0·0001

Moderate problem 71 (11%) 28 (4%) ··

Severe or very severe problem 54 (9%) 32 (5%) ··

Missing 6 (1%) 14 (2%) ··

Household activities

Mean score 17·0 (21·5) 13·2 (20·2) 0·0011

No or a small problem 472 (74%) 564 (80%) 0·039

Moderate problem 91 (14%) 76 (11%) ··

Severe or very severe problem 63 (10%) 55 (8%) ··

Missing 8 (1%) 13 (2%) ··

Mobility activities

Mean score 22·4 (21·1) 17·7 (19·3) <0·0001

No or a small problem 397 (63%) 509 (72%) 0·0023

Moderate problem 146 (23%) 130 (18%) ··

Severe or very severe problem 80 (13%) 63 (9%) ··

Missing 11 (2%) 6 (1%) ··

Social activities

Mean score 15·8 (19·8) 13·4 (18·2) 0·024

No or a small problem 465 (73%) 564 (80%) 0·053

Moderate problem 110 (17%) 93 (13%) ··

Severe or very severe problem 44 (7%) 44 (6%) ··

Missing 15 (2%) 7 (1%) ··

Total score

Mean score 19·1 (18·2) 13·2 (14·9) <0·0001

No or a small problem 448 (71%) 586 (83%) <0·0001

Moderate problem 139 (22%) 94 (13%) ··

Severe or very severe problem 47 (7%) 28 (4%) ··

Data are mean (SD) scores and n (%) for each category. Mean scores per category were compared between groups

using Student’s t test, and distribution of participants across functioning categories was compared with the χ² test.

Lymph-ICF questionnaire=Lymphedema Functioning, Disability, and Health Questionnaire.

Table 2: Patient-reported arm function using the Lymph-ICF questionnaire, 1 year after randomisation,

in the responding population (n=1342)

Articles

www.thelancet.com/oncology Published online August 6, 2024 https://doi.org/10.1016/S1470-2045(24)00350-4 7

To date, the monarchE trial has reported data up to a

median follow-up of 54 months and, considering that

luminal breast cancer can recur late in the disease

course, a treatment effect of abemaciclib on overall

survival might still be found. However, an invasive

disease-free survival benefit does not necessarily

translate into an overall survival benefit.22 In subgroup

analyses of the monarchE trial, patients aged 65 years

and older, those with stage IIB breast cancer, and those

with tumours of Nottingham histological grade 1 had no

significant benefit of treatment with abemaciclib

regarding invasive disease-free survival at a median

follow-up of 42 months (IQR 37–47).6

Hence, in the

subpopulation of SENOMAC used in the current

analysis, of whom 691 (40·5%) of 1705 had stage IIB

breast cancer (in this case, T2N1), 432 (25%) had grade 1

tumours, and 725 (43%) were aged 65 years or older, all

would have gained little, if any, benefit from adjuvant

abemaciclib. Importantly, abemaciclib has a substantially

higher risk of grade 3 or worse adverse events than does

endocrine treatment alone.6,7 Although two treatmentrelated deaths were reported in the abemaciclib plus

endocrine treatment group in monarchE, none occurred

in the endocrine treatment alone group.6

In the

abemaciclib plus endocrine treatment group, 180 (6·4%)

of 2791 patients discontinued both treatments due to

adverse events, while abemaciclib treatment interruption

due to adverse events occurred in 1721 (61·7%) of

2791 patients.

6

To add the toxicity of cALND to this

burden, without the promise of a survival benefit,

appears inappropriate.

The here-selected patient population of the SENOMAC

trial would have no indication for adjuvant abemaciclib if

they had not been found to have pN2–3 status via cALND.

Thus, the breast cancer risk profile of the selected patient

population is probably lower than in the monarchE

population. Recently published results from the

NATALEE trial,8

where cALND was not required to

identify patients with an indication for ribociclib

treatment, might alter the landscape of adjuvant CDK4/6

inhibitors via its broader inclusion criteria.8

However,

these more broad eligibility criteria might also increase

the number of patients receiving the treatment without

gaining any survival benefit.

Several trials have confirmed the non-inferiority of the

omission of cALND compared with sentinel lymph node

biopsy alone with or without nodal radiotherapy.

ACOSOG Z0011 and AMAROS reported equivalent

10-year overall survival rates in 2017 and 2022.1,2 However,

because these trials, as well as the SENOMAC trial,

preceded the approval of abemaciclib by the FDA in

October, 2021,23 and by the European Medicines Agency

in April, 2022,24 abemaciclib was not given to patients in

these trials. Therefore, a potential survival effect of

abemaciclib in the group that received cALND in the

SENOMAC, Z0011, and AMAROS trials could not be

evaluated. The proportion of patients with pN2–3 status

in the cALND groups was 13·7% in ACOSOG Z00112

and

12∙9% in AMAROS,1

which is in line with patients in

SENOMAC who received primary surgery (12·9%).5,25

However, no information on the number of patients with

grade 3 tumours among these patients in Z0011 and

AMAROS is available, which would present an indication

for abemaciclib independent of cALND.

A previous analysis of patient-reported outcome

measures from Swedish and Danish participants in

SENOMAC showed significantly more arm morbidity

1 year after cALND than after sentinel lymph node biopsy

only.13 In the current analysis, updated patient-reported

outcome measures on the here-selected trial population

with a good questionnaire response rate confirm these

findings. Interestingly, although the AMAROS trial

showed an approximately two-times higher risk of

lymphoedema after cALND than after axillary

radiotherapy at 1 year, 3 years, and 5 years of follow-up,1

patient-reported arm symptoms did not significantly

differ between groups at most of these timepoints.1

This

finding might be due to differences in adjuvant

radiotherapy. In SENOMAC, axillary nodal volumes were

irradiated in most patients after cALND, whereas

participants in AMAROS received axillary radiotherapy

after cALND only in case of high nodal burden.1,5 Because

reported arm symptoms and signs of lymphoedema

persisted throughout follow-up in the AMAROS trial,

focusing on preventive measures (eg, restricting the use

of cALND) to reduce the risk for severe arm morbidity

is important.

Our analysis has some limitations. The analysis

evaluating consequences of cALND in the context of the

monarchE trial was not prespecified; however, the

analysis of 1 year patient-reported outcome measures

was prespecified in the study protocol and statistical

analysis plan. Patient-reported outcome measures were

not adjusted for baseline data because they were first

collected after randomisation as an early postoperative

measurement and, thus, a true baseline measure is

absent. However, given the randomised design of the

SENOMAC trial and the large trial population, baseline

arm morbidity should not differ significantly between

groups, and the absence of adjustment for baseline

should not impair the reliability of our 1-year results.

Another limitation is that arm morbidity was exclusively

measured through patient-reported outcome measures

collected by questionnaire because objective measurements of arm volume, circumference, and range of

motion were not part of the study protocol. Patientreported outcome measures without objective measurements might restrict the capture of actual lymphoedema

and restricted range of motion and do not always align

with objective measures. Nevertheless, patient-reported

outcome measures might better reflect the patientexperienced burden than clinician-reported measures.26

Therefore, 5-year arm lymphoedema outcomes reported

in the AMAROS trial1

were also considered. Although

Articles

8 www.thelancet.com/oncology Published online August 6, 2024 https://doi.org/10.1016/S1470-2045(24)00350-4

numbers needed to treat, diagnose, and harm do not

consider patients’ baseline risk, they are point estimates

that are commonly given without a confidence interval.

The absence of a baseline measurement should not be

considered a limitation, especially considering that only

data from randomised trials were used in this analysis.

Thus, our findings offer a simplified but pragmatic

clinical illustration of benefit or harm with a robustness

that is supported by the high number of patients included

in the analysis. Finally, the number needed to treat could

only be based on cohort 1 of the monarchE trial, which

also includes patients with grade 3 and T3 tumours.

Because these patients were not included in the

SENOMAC subpopulation evaluated here, the monarchE

cohort 1 should be regarded as a higher risk population,

indicating that the number needed to treat applicable to

the SENOMAC subpopulation might be even higher

than that calculated here.

Contributors

JdB had the overall responsibility for study design and performance,

funding, data collection, interpretation of data, and writing of the

manuscript. RS did the statistical analysis. JdB and RS directly accessed

and verified the underlying raw data reported in the manuscript.

All authors contributed to study design and data collection, resources,

data curation and project administration, had full access to the data in

the study, critically evaluated and commented on the content of the

manuscript, and assumed final responsibility for the decision to submit

for publication.

Declaration of interests

JdB declares honoraria for educational event lectures for AstraZeneca,

Novartis, and Pfizer. ODG declares honoraria for lectures and

presentations from MSD, AstraZeneca, Eli Lilly, Becton, Dickinson and

Company, and Bayer and the participation on a data safety monitoring or

advisory board for MSD. MS is chairwoman for the Research Committee

at the Swedish Cancer Society. All other authors declare no competing

interests.

Data sharing

De-identified participant data can be made available upon reasonable

request and with the appropriate approvals of relevant Ethical Review

Authorities once the primary endpoint has been published. Proposals

should be directed to the corresponding author at

jana.de-boniface@ki.se; to gain access, data requestors will need to sign

a data access agreement. The study protocol and statistical analysis plan

are available upon request at any time.

Acknowledgments

The SENOMAC trial received funding from the Swedish Research

Council (2015-00760, 2021-02128), the Swedish Cancer Society

(CAN 2015/437, 22 2061 Pj), the Nordic Cancer Union (R217-A13260,

R241-A14982), and the Swedish Breast Cancer Association. The

SENOMAC Trialists’ Group consists of the international steering

committee represented in its entirety in the list of authors. We thank the

Clinical Trials Office at Karolinska University Hospital for data

monitoring and management, the local investigators and study nurses

and other supporting staff at participating hospitals, and the patients

who participated in the trial.

References

1 Bartels SAL, Donker M, Poncet C, et al. Radiotherapy or surgery of

the axilla after a positive sentinel node in breast cancer: 10-year

results of the randomized controlled EORTC 10981-22023 AMAROS

trial. J Clin Oncol 2023; 41: 2159–65.

2 Giuliano AE, Ballman KV, McCall L, et al. Effect of axillary

dissection vs no axillary dissection on 10-year overall survival among

women with invasive breast cancer and sentinel node metastasis:

the ACOSOG Z0011 (Alliance) randomized clinical trial. JAMA

2017; 318: 918–26.

3 Sávolt Á, Péley G, Polgár C, et al. Eight-year follow up result of the

OTOASOR trial: the optimal treatment of the axilla - surgery or

radiotherapy after positive sentinel lymph node biopsy in early-stage

breast cancer: a randomized, single centre, phase III, noninferiority trial. Eur J Surg Oncol 2017; 43: 672–79.

4 Tinterri C, Gentile D, Gatzemeier W, et al. Preservation of axillary

lymph nodes compared with complete dissection in T1-2 breast

cancer patients presenting one or two metastatic sentinel lymph

nodes: the SINODAR-ONE multicenter randomized clinical trial.

Ann Surg Oncol 2022; 29: 5732–44.

5 Donker M, van Tienhoven G, Straver ME, et al. Radiotherapy or

surgery of the axilla after a positive sentinel node in breast cancer

(EORTC 10981-22023 AMAROS): a randomised, multicentre,

open-label, phase 3 non-inferiority trial. Lancet Oncol 2014;

15: 1303–10.

6 Johnston SRD, Toi M, O’Shaughnessy J, et al. Abemaciclib plus

endocrine therapy for hormone receptor-positive, HER2-negative,

node-positive, high-risk early breast cancer (monarchE): results

from a preplanned interim analysis of a randomised, open-label,

phase 3 trial. Lancet Oncol 2023; 24: 77–90.

7 Rastogi P, O’Shaughnessy J, Martin M, et al. Adjuvant abemaciclib

plus endocrine therapy for hormone receptor-positive, human

epidermal growth factor receptor 2-negative, High-risk early breast

cancer: results from a preplanned monarchE overall survival

interim analysis, including 5-year efficacy outcomes. J Clin Oncol

2024; 42: 987–93.

8 Slamon D, Lipatov O, Nowecki Z, et al. Ribociclib plus endocrine

therapy in early breast cancer. N Engl J Med 2024; 390: 1080–91.

9 Harbeck N, Rastogi P, Martin M, et al. Adjuvant abemaciclib

combined with endocrine therapy for high-risk early breast cancer:

updated efficacy and Ki-67 analysis from the monarchE study.

Ann Oncol 2021; 32: 1571–81.

10 Mittendorf EA, King TA, Tolaney SM. Impact of RxPONDER and

monarchE on the surgical management of the axilla in patients with

breast cancer. J Clin Oncol 2022; 40: 3361–64.

11 Williams AD, Ruth K, Shaikh SS, et al. Should patients with

hormone receptor-positive, HER2-negative breast cancer and one

or two positive sentinel nodes undergo axillary dissection to

determine candidacy for adjuvant abemaciclib? Cancer 2024;

130: 1052–60.

12 de Boniface J, Filtenborg Tvedskov T, Rydén L, et al. Omitting

axillary dissection in breast cancer with sentinel-node metastases.

N Engl J Med 2024; 390: 1163–75.

13 Appelgren M, Sackey H, Wengström Y, et al. Patient-reported

outcomes one year after positive sentinel lymph node biopsy with

or without axillary lymph node dissection in the randomized

SENOMAC trial. Breast 2022; 63: 16–23.

14 de Boniface J, Frisell J, Andersson Y, et al. Survival and axillary

recurrence following sentinel node-positive breast cancer without

completion axillary lymph node dissection: the randomized

controlled SENOMAC trial. BMC Cancer 2017; 17: 379.

15 Devoogdt N, Van Kampen M, Geraerts I, Coremans T,

Christiaens MR. Lymphoedema Functioning, Disability and Health

questionnaire (Lymph-ICF): reliability and validity. Phys Ther 2011;

91: 944–57.

16 Devoogdt N, Van Kampen M, Geraerts I, Coremans T,

Christiaens MR. Lymphoedema functioning, disability and health

questionnaire (Lymph-ICF): reliability and validity. Phys Ther 2011;

91: 944–57.

17 Cook RJ, Sackett DL. The number needed to treat: a clinically useful

measure of treatment effect. BMJ 1995; 310: 452–54.

18 Laupacis A, Sackett DL, Roberts RS. An assessment of clinically

useful measures of the consequences of treatment. N Engl J Med

1988; 318: 1728–33.

19 Altman DG. Confidence intervals for the number needed to treat.

BMJ 1998; 317: 1309–12.

20 Johnston SRD, Harbeck N, Hegg R, et al. Abemaciclib combined

with endocrine therapy for the adjuvant treatment of HR+, HER2-,

node-positive, high-risk, early breast cancer (monarchE).

J Clin Oncol 2020; 38: 3987–98.

21 Gaillard T, Piketty J, Feron JG, et al. Rethinking surgical revisions:

impact of the MonarchE trial on axillary dissection in hormonepositive HER2-negative early breast cancer patients potentially

eligible for abemaciclib. Br J Cancer 2024; 130: 1141–48.

Articles

www.thelancet.com/oncology Published online August 6, 2024 https://doi.org/10.1016/S1470-2045(24)00350-4 9

22 Meirson T, Goldstein DA, Gyawali B, Tannock IF. Review of the

monarchE trial suggests no evidence to support use of adjuvant

abemaciclib in women with breast cancer. Lancet Oncol 2023;

24: 589–93.

23 Royce M, Osgood C, Mulkey F, et al. FDA approval summary:

abemaciclib with endocrine therapy for high-risk early breast

cancer. J Clin Oncol 2022; 40: 1155–62.

24 European Medicines Agency. Verzenios (abemaciclib). Amsterdam:

2022. https://www.ema.europa.eu/en/documents/overview/

verzenios-epar-medicine-overview_en.pdf (accessed

March 16, 2024).

25 Giuliano AE, Hunt KK, Ballman KV, et al. Axillary dissection vs no

axillary dissection in women with invasive breast cancer and

sentinel node metastasis: a randomized clinical trial. JAMA 2011;

305: 569–75.

26 Sackey H, Magnuson A, Sandelin K, et al. Arm lymphoedema after

axillary surgery in women with invasive breast cancer. Br J Surg

2014; 101: 390–97.