Part XI · Hormone receptor-positive, HER2-negative disease · Chapter 60

Adjuvant endocrine therapy

Why most rows in these tables answer a question one step to the side of the one the clinic is asking.

1 · Scope, and how to read the evidence tables in this chapter

This chapter is the evidence compendium for adjuvant endocrine therapy. It collects the randomised trials that settled which agent, for how long, and with or without ovarian suppression. It ends on adherence, because that is the largest modifiable determinant of how much of the benefit is actually delivered.

Three neighbouring chapters carry what is deliberately absent from this one. The pharmacology of each agent class is in Endocrine therapy. The decision architecture for early hormone receptor-positive disease, including the genomic assays and the chemotherapy question, is in Adjuvant chemotherapy and the genomic assay decision. Adjuvant CDK4/6 inhibition is a separate decision and has its own chapter in Adjuvant CDK4/6 inhibition.

The table below is a filter on registry fields rather than on what a trial randomised. That distinction does more work here than anywhere else in the book, because endocrine therapy was a component of nearly every adjuvant trial run in this subtype for forty years. A trial enters the table if endocrine therapy was anywhere in its design. It did not have to be the variable.

Sorted by what was actually randomised, the thirty-two rows fall into six groups. Five of them are the sections that follow: tamoxifen duration, aromatase inhibitor against tamoxifen, ovarian suppression, extension beyond five years, and the oral degraders. The sixth group has no section of its own, and its rows are worth naming so that none of them goes unexplained.

Four rows sit in the table because endocrine therapy was the constant rather than the variable. NSABP B-20 and SWOG-8814 / INT-0100 added chemotherapy to tamoxifen. WSG-PlanB randomised two chemotherapy regimens. Its assay-driven omission of chemotherapy sat outside that randomisation. WSG-ADAPT HR+/HER2− compared two assay-defined groups who both received endocrine therapy alone, which is a selection rule tested for non-inferiority rather than a randomised drug comparison. All four belong to Genomic assays and the decision to omit chemotherapy, and none of them randomised an endocrine decision.

They do carry one thing the endocrine sections cannot. The 21-gene recurrence score was built and validated in cohorts whose endocrine therapy was tamoxifen alone^31,32. The assay now used to place a patient on endocrine therapy without chemotherapy was calibrated against a backbone most of those patients no longer receive. That is a measurement problem rather than a biological one, and Clinical and genomic risk assessment is where it is argued.

NSABP B-14 is the oldest row and the one the rest of the chapter stands on. It randomised tamoxifen against placebo in node-negative receptor-positive disease²⁵. No trial since has carried a no-treatment arm in this population. Every comparison below is therefore between two active strategies, and the absolute benefit of endocrine therapy itself is measured nowhere in this chapter.

Two rows randomised nothing that bears on a drug choice. CALGB 9343 randomised radiotherapy in women aged seventy or older, with tamoxifen in both arms, and belongs to Omission in low-risk and older patients²⁷. POSITIVE is a single-group study of a planned interruption to attempt pregnancy, with its comparison drawn from an external cohort rather than from randomisation²⁸. It is discussed in Pregnancy after breast cancer and interrupting endocrine therapy.

The last two rows are the within-class question, and this section rather than the next one carries them because the filter below is built on the comparison against tamoxifen. FACE randomised letrozole against anastrozole in node-positive disease. MA.27 randomised exemestane against anastrozole under a two-sided superiority design. Neither trial separated the drugs on its primary endpoint. FATA-GIM3 randomised across the same three drugs inside a larger design and also separated them on nothing⁸. Three randomised comparisons agree that the choice among the three aromatase inhibitors is not an efficacy choice. It is a tolerability and interaction choice, and it should be made on that basis^12,16.

What none of the three can show is the manoeuvre a clinic makes after an arthralgia. Nobody randomised a switch between aromatase inhibitors against continuing the first one. Moving a symptomatic patient from letrozole to exemestane therefore rests on the absence of an efficacy difference and on nothing else.

The overview table prints no result column at all, which is deliberate. Outcomes belong to the per-question tables below, where the control arm is stated next to them. Read the arms column of any generated table in this book before reading anything else in it.

Several of the sections below open with the question their trials were built to answer, taken from the [Chapter Stories](stories.html) map, where the whole landscape is laid out as questions. The division of labour is that tables enumerate and prose argues. A figure that sits in a table is not repeated in the text unless the argument is about that figure.

2 · Tamoxifen and the trials that settled duration

The duration question is not about how long the drug works. It is about how long its effect persists after it stops.

The table holds two rows and one result. ATLAS randomised continuation to ten years against stopping at five, and reported. aTTom sits in the registry as awaited, with no primary duration result on its record and its primary endpoint not stated there. The ten-against- five decision therefore rests on a single trial. The pooled reading of the two that circulates in review articles is an arithmetic performed outside either trial's own publication.

The accounting matters more than the size of the result. ATLAS counts recurrence and breast cancer mortality over years 5 to 14¹⁵. That window runs five years past the last tablet in the continuation arm. The hazard of recurrence in this subtype does not fall away when treatment stops³³. A duration trial that closed its books at year ten would be reading the effect while it was still arriving.

That is the whole case for extending anything. It is the mirror image of the late recurrence hazard curve set out in Annual hazard of recurrence and the long tail of hormone receptor-positive disease. A treatment whose effect arrives late is worth giving only to a woman whose risk also arrives late.

Two features of the population constrain how the figures transfer. ATLAS randomised women already free of disease at five years, so the estimate is conditional on surviving that far. Its breast cancer outcomes are reported only in the women with ER-positive disease, and the remainder were ER-negative or never assayed, which reflects practice between 1996 and 2005¹⁵.

The price of the second five years is mostly endometrial and thrombotic. Continuing to ten years raised endometrial cancer by a rate ratio of 1.74 and pulmonary embolus by 1.87, each against stopping at five¹⁵. Both are ratios of event rates. The absolute excess depends on the baseline incidence in the woman being counselled, which a ratio does not carry.

What the table cannot show is the comparison a postmenopausal woman is actually offered. ATLAS set continuation against stopping. Nobody randomised ten years of tamoxifen against five years of tamoxifen followed by an aromatase inhibitor, which is the sequence the next two sections make the default. The one analysis that does sit on the aTTom record asks instead whether an expression assay identifies the women who gain²⁹, and Unresolved questions and what reads out next returns to it.

The patient-level overview of the first five years remains the anchor, and it established that a carry-over effect exists at all³⁴. The earlier overviews had already shown that tamoxifen reduces recurrence and mortality in ER-positive and ER-unknown disease, and that the mortality reduction keeps growing to 15 years^35,36.

HER2 was once proposed as a marker of tamoxifen resistance^37,38,39. In adjuvant trials the benefit of tamoxifen did not depend on HER2 or p53 status^40,41. In metastatic disease, HER2-positive tumours responded less to every endocrine treatment in a meta-analysis⁴², and HER2 amplification with HER1 expression marked lower ER levels and a modestly smaller tamoxifen response⁴³. Quantitative measurement of ErbB-2 carried prognostic and predictive information of its own⁴⁴. In TransATAC, low ER or PgR and high HER2 marked a higher recurrence risk with either drug, and anastrozole was better than tamoxifen in every molecular subgroup⁴⁵. Tamoxifen is still the first choice in a premenopausal woman who will not receive ovarian suppression, and in anyone for whom bone loss is the dominant competing risk.

3 · Aromatase inhibitors against tamoxifen, and the menopausal status that decides

Menopausal status is not a patient characteristic in this decision. It is a statement about where oestrogen is being made, and therefore about which blockade will work.

In postmenopausal women an aromatase inhibitor outperforms tamoxifen. A patient-level meta-analysis of 31,920 women found that recurrence rate ratios favoured aromatase inhibitors while the treatments differed, at 0.70 with a 95% confidence interval of 0.64 to 0.77, and not significantly thereafter. Ten-year breast cancer mortality across all comparisons was reduced, with a rate ratio of 0.86 and an interval of 0.80 to 0.94⁴⁶. The same analysis priced the trade. Ten-year endometrial cancer incidence was 0.4% against 1.2% with tamoxifen, and five-year bone fracture risk was 8.2% against 5.5%.

In premenopausal women the ovary is the dominant source, so an aromatase inhibitor alone does not work. It requires ovarian suppression, and the comparison then changes character. A patient-level meta-analysis of 7,030 women receiving ovarian suppression found that an aromatase inhibitor reduced breast cancer recurrence against tamoxifen, with a rate ratio of 0.79 and an interval of 0.69 to 0.90. The absolute reduction in five-year recurrence risk was 3.2 percentage points, from 10.1% to 6.9%. Breast cancer mortality did not differ, with a rate ratio of 1.01 and an interval of 0.82 to 1.24. Fractures were more common, at 6.4% against 5.1%⁴⁷. The two estimates do not have the same shape. Mortality fell alongside recurrence in the postmenopausal comparison and did not under suppression, while a named toxicity rises in both. The agents themselves are covered in Endocrine therapy and the suppression trials in Ovarian function suppression and who benefits.

The six randomised trials underneath the postmenopausal estimate agree on one thing, and it is easy to lose in the detail. Wherever tamoxifen exposure is replaced by aromatase inhibition, recurrence falls. Wherever both arms finish on an aromatase inhibitor and only the order differs, nothing separates them. That is one finding about the class rather than six findings about six protocols.

The six divide by control arm rather than by drug. ATAC and BIG 1-98 compared five years of an aromatase inhibitor against five years of tamoxifen from the start. IES and ABCSG-8 / ARNO 95 compared a switch at two to three years against continuing tamoxifen to the end of five. TEAM and FATA-GIM3 compared starting with an aromatase inhibitor against starting with tamoxifen and switching later. Three different questions, habitually read as one.

The component reports are the primary evidence under the meta-analysis. ATAC favoured anastrozole for recurrence at its first analysis, at the completion of treatment and at 100 months^48,49,50. BIG 1-98 first reported in 2005⁵¹. Its sequencing comparison found that neither sequence improved disease-free survival over letrozole alone⁵². IES reported its first switch result in 2004⁵³. ARNO 95 found improved overall survival after switching to anastrozole⁵⁴, and the Italian ITA trial improved event-free and recurrence-free survival^55,56. A meta-analysis of the anastrozole switch trials found an overall survival benefit⁵⁷. A patient-level meta-analysis of the aromatase inhibitor trials found lower recurrence whether the drug was given from the start or after two to three years of tamoxifen⁵⁸.

The same trials measured what the benefit costs. In ATAC, patients with normal bone density at the start did not appear to need monitoring beyond that recommended for healthy postmenopausal women⁵⁹. Endometrial abnormalities were fewer with anastrozole⁶⁰. Overall quality of life was similar, with different symptom profiles⁶¹. Prior chemotherapy did not change the relative benefit⁶². Tamoxifen lowered anastrozole blood levels without affecting oestradiol suppression⁶³. In BIG 1-98, letrozole increased bone fractures⁶⁴, and cardiovascular events were infrequent and differed between the arms⁶⁵.

TEAM constrains practice more than the others and is cited least. Its null result means that starting with an aromatase inhibitor is not superior to starting with tamoxifen and switching later¹⁸. FATA-GIM3 reached the same place from a six-arm design⁸. The case for starting with an aromatase inhibitor therefore rests on tolerability and on the gynaecological and thromboembolic profile. It does not rest on efficacy, because the two trials built to test the order found none.

The decision rests on BIG 1-98. It was double-blind. It randomised the whole five years rather than a switch part-way. It is also the only one of the six whose intention-to-treat overall survival comparison favoured the aromatase inhibitor. ATAC, with ten years of follow-up, did not reproduce that¹⁹. The divergence is not anastrozole against letrozole, because FACE randomised those two head to head and separated them on nothing. It is blinding, crossover and duration of follow-up.

BIG 1-98 also shows how much of its own survival claim is model-dependent. After the 2005 analysis, patients still on tamoxifen were unblinded and offered letrozole. A censoring-weighted analysis that removes those crossovers moves the overall survival estimate further in favour of letrozole than the intention-to-treat estimate in the table¹⁷. Both are correct estimates of different quantities. Only the intention-to-treat estimate describes the policy that was actually studied.

The two switch trials differ from each other in a way the table hides. IES randomised inside a single protocol with a fixed five-year total and reported overall survival²². The ABCSG-8 and ARNO 95 row is a combined analysis of two trials, reporting event-free survival with no overall survival in the pooled paper²⁶. Their agreement is real. It is agreement between one trial and one pooling, not between two replications.

The measurement point is the one that causes harm in clinic. Menopausal status after chemotherapy is inferred from amenorrhoea and from gonadotrophin and oestradiol assays. Amenorrhoea is not menopause. An aromatase inhibitor given to a woman with recoverable ovarian function lowers oestrogen, raises gonadotrophins and can restart the ovary. In an audit of 45 women with chemotherapy-induced amenorrhoea who started an aromatase inhibitor, 27% showed a return of ovarian function, and the authors advised caution with sensitive oestradiol monitoring⁶⁶. Markers of ovarian reserve do not settle it either. Chemotherapy lowers Mullerian inhibiting substance rapidly, and the fall did not predict later menstrual function⁶⁷.

Three things the table cannot show. Every one of the six enrolled postmenopausal women, so none of it transfers to a premenopausal woman who is not suppressed. No trial randomised the length of the tamoxifen lead-in, so the two to three years the switch trials used is a protocol convention rather than an optimum. And bone is not a column here, although the meta-analysis above prices recurrence, mortality and fracture together.

4 · Ovarian function suppression and who benefits

Four rows, three questions, and only two of these trials randomised suppression at all. SOFT set tamoxifen plus suppression, and exemestane plus suppression, against tamoxifen alone. ASTRRA set two years of goserelin added to tamoxifen against tamoxifen alone. SOFT/TEXT and ABCSG-12 gave a gonadotropin-releasing hormone agonist to everybody and randomised the partner drug instead. A reader who takes all four as tests of ovarian suppression has read two of them backwards.

Where the four agree is narrower than it looks. Adding suppression to tamoxifen lowers recurrence in premenopausal hormone receptor-positive disease, and in ASTRRA it also moved overall survival. Upgrading the partner drug on top of suppression is a separate claim, and the two trials that tested it disagree.

The decision rests on SOFT, because it is the only trial here that randomised the addition against the thing a clinic would otherwise give, which is five years of tamoxifen. Its population was not pre-selected on recovered ovarian function, which is what makes the row load-bearing. Its first report did not find the addition significant⁶⁸. The eight-year report reversed that reading on disease-free and overall survival alike¹⁰, and the twelve-year report held it⁶⁹. The final report, at a median of fifteen years, kept the reduction in breast cancer recurrence and lost the survival difference¹. No biology changed between the first report and the second. The events needed time to accrue, which is the structural problem with adjuvant endpoints set out in Neoadjuvant, adjuvant, and metastatic design compared.

ASTRRA asks the same addition question in a narrower population. It enrolled women aged 45 or younger who had already had chemotherapy and remained premenopausal or resumed ovarian function, confirmed every six months by follicle-stimulating hormone and bleeding history⁵. Two years of goserelin improved disease-free survival, and the separation persisted at eight years⁷⁰. That population is a prospective enrichment of precisely the SOFT stratum in which suppression looked worthwhile, so its agreement with SOFT is close to circular. What it adds is that the effect survives being looked for on purpose. Its overall survival comparison did reach significance, and the interval around that estimate is wide enough that the point estimate should not be quoted to a patient as a survival benefit.

On the partner question the two trials disagree, and the disagreement is not about the drugs. SOFT/TEXT found exemestane better than tamoxifen on disease-free survival and not on overall survival, at twelve years and again at fifteen^71,1. ABCSG-12 found anastrozole no better than tamoxifen. Both held suppression constant in every arm, so that is not what divides them. Four other things do. ABCSG-12 gave three years of treatment rather than five. It used anastrozole rather than exemestane. It enrolled a largely chemotherapy-naive and lower-risk population, and its primary report came before most late events in this subtype have occurred²⁰. And its final analysis was powered around a second randomisation, to zoledronic acid, which belongs to the adjuvant bisphosphonate argument rather than to this one⁷². Its endocrine comparison is a marginal one inside a bone-agent trial, and the patient-level meta-analysis that pools the field averages that result into the total⁴⁷. Reading the two rows as a verdict on anastrozole against exemestane is the error the arms column exists to prevent.

The subgroups are where the decision lives, and the table carries only the HER2-negative hazard ratio. In HER2-negative disease, 86.0% of the joint population, the 15-year overall survival gain from exemestane over tamoxifen, both with suppression, was larger in women who received chemotherapy. It was 3.9 percentage points in SOFT and 2.5 in TEXT. Young age and high grade also marked a larger gain¹. The gain tracks baseline risk rather than the drug, so a single average for this comparison is not usable in clinic.

One inversion inside SOFT is worth more attention than it gets. Exemestane with suppression gave the best eight-year disease-free survival of the three arms and did not give the best eight-year overall survival¹⁰. The arm that beat tamoxifen alone on survival was tamoxifen with suppression. The trial was not powered to rank the two suppression arms against each other on survival. Read the recurrence ordering as a survival ordering and you are reading something the design did not test. The final report removed the inversion without settling the ranking, because no arm differed significantly from tamoxifen alone on overall survival¹.

Three things nobody randomised are missing from every row, and they are the part of the decision that is hardest in clinic. Duration is one, at two years in ASTRRA, three in ABCSG-12 and five in SOFT and TEXT, never varied inside a trial. Method is the second. ABCSG-12 and ASTRRA specified a gonadotropin-releasing hormone agonist, and no trial here set one method against another, so oophorectomy and radiation ablation are inferred rather than tested. The third is confirmation of suppression. ASTRRA measured ovarian function repeatedly and the others did not, which means the treatment being compared is biochemically the same drug and not reliably the same state. Discontinuation is not a column either, although it is what decides how much of any of this is delivered.

5 · Extended adjuvant therapy and the shape of the late benefit

The case for extension is a statement about hazard. Distant recurrence in this subtype accrues at a roughly constant rate from year 5 to year 20³³. The size of that residual risk by tumour size and nodal status is tabulated in Annual hazard of recurrence and the long tail of hormone receptor-positive disease and Late recurrence risk and surveillance implications. Extension is therefore a dormancy intervention rather than consolidation of a falling risk, and Dormancy, residual disease, and late recurrence is where that argument is made properly.

One thing is worth settling before the rows are read. Four of the eight report overall survival, and none of those four shows an advantage for the longer arm. The other four do not report it at all. An extension decision is a recurrence decision, and it should be described to a patient as one.

The eight look inconsistent until they are sorted by control arm, which is the axis that decides what each can answer. MA.17, MA.17R and NSABP B-42 randomised further treatment against placebo, and they are the only three that bear on whether to extend at all. ABCSG-16, DATA, GIM4 and IDEAL randomised a longer extension against a shorter one, so they answer how long rather than whether. SOLE answers neither, because both of its arms were extended.

The three placebo-controlled trials answer whether to extend at all, and they answer it in both sequences. MA.17 extended letrozole after tamoxifen. MA.17R extended it after an aromatase inhibitor, and both separated from placebo on disease-free survival^23,14. NSABP B-42 asked the same question in the same place as MA.17R and did not clear its own significance threshold⁶. Extending after an aromatase inhibitor is therefore not the weaker version of the question, which is how it is usually read.

The four duration-against-duration trials are where the field reads a disagreement that may not be there. DATA and GIM4 asked the same thing, a longer aromatase inhibitor course after two to three years of tamoxifen, and returned disease-free survival hazard ratios of 0.79 and 0.78. Only GIM4 crossed significance, and it read out at a median follow-up of 11.7 years against DATA's five-year adapted endpoint^11,4. What separates a positive extension trial from a null one here is events and follow-up rather than drug or sequence. IDEAL admitted women after any five years of endocrine therapy, so each of its arms mixes the two sequences⁹.

NSABP B-42 is where a convention rather than a biology decided the reported verdict. Its disease-free survival p value fell below 0.05 and above the threshold its own design had prespecified, so the trial is recorded as negative⁶. The ten-year update reached significance on the same endpoint⁷³. Nothing about letrozole changed between those reports. An interim-analysis alpha changed what the same comparison was permitted to be called, which is worth knowing before quoting either one.

ABCSG-16 is the only row that priced the longer course in fractures as a named secondary endpoint rather than as an adverse event count³. Its two arms differed only in how many further years of anastrozole were given. The fracture excess is therefore attributable to duration and not to starting the drug, which is the one place in this table where the cost of extension is reported as an outcome.

SOLE answers a question the others do not ask. It gave letrozole intermittently against continuously, on the hypothesis that a break restores sensitivity, and found no difference at five years or at seven^13,74. How to give an extension is separate from whether to give one. A null schedule trial is not evidence about duration.

The reading that fits is about what the control arm was doing, not about the molecule or the sequence. Against no further treatment, the longer arm lowered recurrence in MA.17 and MA.17R, in neither case with a survival difference. Against a shorter further course, three of the four trials did not separate. MA.17R is the awkward row for any tidy account of diminishing returns. Its patients had already had about five years of an aromatase inhibitor, and most had taken tamoxifen before that.

Its later reports show the same benefit through the crossover. The updated analysis at a median of 30 months confirmed better disease-free and distant disease-free survival with letrozole⁷⁶. Analyses adjusted for the crossover favoured letrozole for disease-free, distant disease-free and overall survival, with an overall survival hazard ratio of 0.61 by one method and 0.76 by another⁷⁷. In a non-randomised cohort analysis, women who started letrozole after a gap since tamoxifen also benefited⁷⁸. The price was a modest fall in bone density⁷⁹ and more vasomotor symptoms, without an adverse effect on overall quality of life⁸⁰. ABCSG-6a gave three years of anastrozole after five of tamoxifen and reduced recurrence against no further treatment⁸¹.

The patient-level meta-analysis of twelve aromatase inhibitor extension trials in 22,031 postmenopausal women who had already completed at least five years of endocrine therapy now prices both sides⁸². Allocation to further aromatase inhibition reduced recurrence, with a rate ratio of 0.73 and a 95% confidence interval of 0.67 to 0.80. What a clinic needs from it is the absolute framing rather than the proportional one, and that arithmetic is in Absolute and relative benefit, and why the difference matters.

The subset that matters most in current practice is the one already treated with an aromatase inhibitor. There, five further years reduced recurrence with a rate ratio of 0.71, taking the risk from year 5 to year 15 from 15.2% to 11.6%. Distant recurrence fell with a rate ratio of 0.73, from 8.6% to 6.6%. Breast cancer mortality fell non-significantly, with a rate ratio of 0.90, an interval of 0.70 to 1.15 and a p value of 0.40, from 5.0% to 4.4%⁸².

Because the proportional reduction was similar across tumour characteristics, the absolute benefit scales with baseline risk. Ten years against five years took node-positive risk from 20.1% to 16.3% and node-negative risk from 11.8% to 9.1%⁸². Five-year fracture risk rose with a rate ratio of 1.35, from 3.4% to 4.6%.

One component of the extension benefit is not recurrence at all. Contralateral and second primary breast cancers fall with continued aromatase inhibition, and IDEAL isolates that effect because its two arms differed only in duration⁹. MA.17R reports the same direction on annual contralateral incidence¹⁴. Preventing a new primary is a real benefit. It is not the same achievement as preventing relapse, and Treatment duration and extended therapy sets out why the distinction matters when quoting a disease-free survival difference.

Two populations are missing from every row. No extension trial enrolled premenopausal women, so duration in a woman treated with tamoxifen and suppression is unstudied. And every placebo- controlled trial enrolled women who had already tolerated five years, which selects for adherence and against the patients most likely to stop. No trial in the table allocated extension on a molecular signal, which is the gap Unresolved questions and what reads out next returns to.

6 · Late recurrence risk and surveillance implications

This subtype recurs on a timescale that outlasts most follow-up schedules and most trials. The annual hazard of recurrence does not fall to zero after five years, and the pattern across 24 years of follow-up is set out in⁸³ and⁸⁴.

The size of that residual risk is now precisely quantified. Among 62,923 women with oestrogen receptor-positive disease who were disease-free after five years of scheduled endocrine therapy, recurrences occurred at a steady rate from year 5 to year 20. Distant recurrence risk over that window was 13% for T1N0 disease, 20% for T1N1-3, 34% for T1N4-9, 19% for T2N0, 26% for T2N1-3 and 41% for T2N4-9³³.

Grade added moderately within a stage stratum. For T1N0 disease the risk over years 5 to 20 was 10% for low-grade, 13% for moderate-grade and 17% for high-grade tumours³³. Progesterone receptor and HER2 status were not predictive of late distant recurrence in that dataset.

Two implications follow, and only one of them is comfortable.

The comfortable one is that the original pathology report retains predictive power for two decades. Size and nodal status are not superseded by anything currently in routine use.

The uncomfortable one is that no surveillance strategy has been shown to convert that knowledge into benefit. Detecting asymptomatic distant recurrence earlier has not improved survival, and the evidence against surveillance intensity is in Rehabilitation and survivorship. Circulating tumour DNA detects molecular relapse before imaging does, and whether acting on that lead time helps is unresolved and is the subject of Liquid biopsy and longitudinal monitoring. Late recurrence is a dormancy problem, and its biology is in Dormancy, residual disease, and late recurrence.

7 · Oral selective estrogen receptor degraders in the adjuvant setting

No adjuvant oral selective estrogen receptor degrader has reported. The table carries three rows and no outcome behind any of them. Everything in this section is a hypothesis with a trial design attached to it.

The rationale was transposed from advanced disease, where the target is an ESR1 ligand-binding domain mutation conferring ligand-independent receptor activity. The biology is in Estrogen receptor signaling and endocrine resistance and the agents are in Oral degraders and next-generation agents. The transposition has a problem that is rarely stated. ESR1 mutations are selected by aromatase inhibitor exposure and are uncommon at diagnosis Endocrine therapy and the emergence of ESR1 alterations. In the adjuvant setting there is no measurable tumour, and usually no mutation to target.

The adjuvant hypothesis therefore cannot be that a degrader treats a mutant receptor. It has to be that more complete receptor blockade stops resistance from forming. That is a different claim and it has never been tested in any setting.

The CAMBRIA pair tests two versions of it, split by when the degrader is given rather than by which degrader it is. CAMBRIA-1 switches patients to camizestrant after two to five years of standard endocrine therapy, which is the late-selection version. CAMBRIA-2 replaces standard endocrine therapy from the start, which is the complete-blockade version. lidERA puts a second molecule into the same question, giredestrant against standard adjuvant endocrine therapy. Its registry record does not yet fix a primary endpoint, so what it will be judged on is not settled here. Two agents behaving the same way would say something about the strategy. One of each would say something about the molecules.

Two design features deserve attention now rather than at readout. Invasive breast cancer-free survival counts second primaries, so part of any positive result will be prevention rather than recurrence control, exactly as in the extension trials above. And CAMBRIA-1 admits patients who received a CDK4/6 inhibitor, while CAMBRIA-2 permits abemaciclib in both arms for its first two years³⁰. The comparator in each is current practice rather than endocrine monotherapy.

What no trial in the class does is select. None of the three randomises on circulating tumour DNA, on ESR1 status or on any measure of residual disease. A positive trial would therefore license a drug for a population in which most patients were never going to relapse. The interception argument is in Liquid biopsy and longitudinal monitoring and Strategies to detect and intercept residual disease.

8 · Adherence, toxicity, and the gap between prescribed and taken

Every hazard ratio in this chapter is an intention-to-treat estimate over a population in which a large minority stopped taking the drug. That is not a footnote to the effect size. It is part of what the effect size means, and the arithmetic follows from it.

In a cohort of 8,769 women with stage I to III hormone-sensitive breast cancer who filled at least one prescription, 32% had discontinued endocrine therapy by 4.5 years. Among those who continued, 72% were fully adherent by a medication possession ratio of 80% or more. Combining the two, 49% took adjuvant endocrine therapy for the full duration on the intended schedule. Women under 40 had the highest discontinuation risk, with a hazard ratio of 1.51 and a 95% confidence interval of 1.23 to 1.85⁸⁵.

Compare that against the size of the other lever a clinician controls. Choosing an aromatase inhibitor over tamoxifen in a postmenopausal woman changes recurrence by a few percentage points over a decade. Not taking the tablet removes the entire effect of the drug. Clinic time is spent almost entirely on the smaller lever.

The randomised evidence says most of that loss is not pharmacological. In the placebo-controlled extended-therapy trials pooled by the EBCTCG, non-adherence was 39.0% on further aromatase inhibition and 37.6% on placebo⁸². Almost as many women stopped an inert tablet as stopped an active one, which says that the burden is the regimen as much as the pharmacology. Symptom management therefore has a ceiling, and the rest of the problem is structural.

The toxicities that drive this are quantified elsewhere in the chapter. Fractures rise with aromatase inhibition in postmenopausal women⁴⁶ and with ovarian suppression in premenopausal women⁴⁷. Endometrial cancer and thrombosis rise with tamoxifen⁴⁶. Arthralgia and vasomotor symptoms are the commonest reasons given for stopping, and the symptom burden is developed in Endocrine therapy and Rehabilitation and survivorship.

Measurement matters here more than usual, because the three common methods disagree by design. Self-report overestimates, and most of all when the prescriber is the one asking the question. Pharmacy refill measures collection from a dispensary rather than ingestion. Serum tamoxifen or endoxifen, and serum oestradiol during aromatase inhibition, measure exposure directly and are almost never used. A trial reporting adherence by refill and a clinic reporting it by self-report are not describing the same quantity, and neither is wrong.

Dose is the one pharmacological lever with randomised evidence behind it. TAM-01 tested 5 mg of tamoxifen daily against placebo and reported durable benefit at ten years⁸⁶. That trial enrolled breast intraepithelial neoplasia rather than invasive disease. Using low-dose tamoxifen to rescue an intolerant patient with invasive cancer is an extrapolation, and it should be described to her as one.

9 · Unresolved questions and what reads out next

The filter above asks for unreported trials in early disease with an endocrine or cell-cycle component. It returns three adjuvant phase 3 trials and one neoadjuvant phase 2 with no control arm. Read the line and phase columns before treating the four as one programme.

Five questions are open, and only two of them have a trial that will answer them.

Whether an oral degrader beats standard endocrine therapy in the adjuvant setting is being tested in two positions at once. CAMBRIA-1 and CAMBRIA-2 could disagree. A split result would be more informative than two concordant ones. A positive switch trial with a negative upfront trial would say the benefit lies in late blockade, after resistance has begun to form.

Whether extension can be targeted rather than offered broadly has been tested only retrospectively. The Breast Cancer Index began as a two-gene expression ratio, HOXB13 to IL17BR, associated with outcome under tamoxifen⁸⁹. In TransATAC its linear model was the only one of three tests prognostic for both early and late distant recurrence⁹⁰. In tumours from IDEAL, a high ratio predicted benefit from five rather than two and a half further years of letrozole, with a hazard ratio of 0.42, and a low ratio did not, at 0.95⁹¹. The interaction p value was 0.045. The same kind of analysis was run in aTTom. In its first report, from 583 node-positive patients, a high ratio predicted benefit from ten rather than five years of tamoxifen, with a hazard ratio of 0.35⁹². A low ratio did not, at 1.07. The final analysis on the aTTom record found the same split in its node-positive patients²⁹. These are predictive claims made inside trials randomised for something else, which is where the 21-gene score stood before TAILORx Genomic assays and the decision to omit chemotherapy. Nobody has randomised extended endocrine therapy on a molecular signal, and circulating tumour DNA is the other obvious candidate. The assay performance and the interception problem are in Liquid biopsy and longitudinal monitoring and Strategies to detect and intercept residual disease.

Duration in premenopausal women is close to unstudied. Every aromatase inhibitor extension trial in this chapter required a postmenopausal population. ATLAS carried no such requirement, and it tested tamoxifen rather than the suppression-based regimens these women now receive.

The duration of ovarian suppression has never been compared head to head in a randomised trial. ASTRRA used two years and SOFT used five. The question is not academic, because suppression is the least tolerated component of the regimen in young women Young adults, fertility, and pregnancy.

Whether cell cycle inhibition changes the endocrine question is being approached from two directions here. WSG ADAPTcycle asks whether ribociclib with an aromatase inhibitor can replace chemotherapy in intermediate-risk disease, with entry decided by a recurrence score and by Ki-67 measured after short preoperative endocrine therapy⁸⁷. A positive result would move part of the chemotherapy decision in Genomic assays and the decision to omit chemotherapy into the endocrine column. That is a larger change than any degrader result would produce.

Its entry rule is also the clearest example in this chapter of a measurement standing in for a biology. Endocrine response is defined by a Ki-67 threshold, scored locally in some centres and centrally in others, and the two readings do not agree in every patient⁸⁷. Allocation therefore depends partly on which laboratory read the slide. The scoring conventions are set out in Ki67, PEPI and response measures for a setting where pCR fails.

NEOLETRIB is the odd row and the one worth keeping. It is a single-group neoadjuvant study of letrozole with ribociclib, with serial biopsy and single-cell sequencing during treatment⁸⁸. It will not settle a treatment question and is not built to. It is the only study in this chapter that samples a tumour while endocrine pressure is being applied, which is the measurement the adjuvant setting can never make. The window designs it belongs among are in Neoadjuvant CDK4/6 inhibition and window-of-opportunity designs.

References

All 92 references

1. Francis PA, et al. Final outcomes of the SOFT and TEXT phase III trials in premenopausal hormone receptor-positive early breast cancer. Ann Oncol 2026. PMID 42229584
2. Nitz UA, et al. Endocrine Therapy Response and 21-Gene Expression Assay for Therapy Guidance in HR+/HER2- Early Breast Cancer. J Clin Oncol 2022 40:2557-2567. PMID 35404683
3. Gnant M, Fitzal F, Rinnerthaler G, et al. Duration of adjuvant aromatase-inhibitor therapy in postmenopausal breast cancer. N Engl J Med 2021 385:395-405. PMID 34320285
4. Del Mastro L, et al. Extended therapy with letrozole as adjuvant treatment of postmenopausal patients with early-stage breast cancer: a multicentre, open-label, randomised, phase 3 trial. Lancet Oncol 2021 22:1458-1467. PMID 34543613
5. Kim HA, Lee JW, Nam SJ, et al. Adding ovarian suppression to tamoxifen for premenopausal breast cancer: a randomized phase III trial. J Clin Oncol 2020 38:434-443. PMID 31518174
6. Mamounas EP, Bandos H, Lembersky BC, et al. Use of letrozole after aromatase inhibitor-based therapy in postmenopausal breast cancer (NRG Oncology/NSABP B-42): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet Oncol 2019 20:88-99. PMID 30509771
7. Nitz U, et al. West German Study PlanB Trial: Adjuvant Four Cycles of Epirubicin and Cyclophosphamide Plus Docetaxel Versus Six Cycles of Docetaxel and Cyclophosphamide in HER2-Negative Early Breast Cancer. J Clin Oncol 2019 37:799-808. PMID 30785826
8. De Placido S, et al. Adjuvant anastrozole versus exemestane versus letrozole, upfront or after 2 years of tamoxifen, in endocrine-sensitive breast cancer (FATA-GIM3): a randomised, phase 3 trial. Lancet Oncol 2018 19:474-485. PMID 29482983
9. Blok EJ, Kroep JR, Meershoek-Klein Kranenbarg E, et al. Optimal duration of extended adjuvant endocrine therapy for early breast cancer; results of the IDEAL trial (BOOG 2006-05). J Natl Cancer Inst 2018 110. PMID 28922787
10. Francis PA, Pagani O, Fleming GF, et al. Tailoring adjuvant endocrine therapy for premenopausal breast cancer. N Engl J Med 2018 379:122-137. PMID 29863451
11. Tjan-Heijnen VCG, van Hellemond IEG, Peer PGM, et al. Extended adjuvant aromatase inhibition after sequential endocrine therapy (DATA): a randomised, phase 3 trial. Lancet Oncol 2017 18:1502-1511. PMID 29031778
12. Smith I, et al. Comparative Efficacy and Safety of Adjuvant Letrozole Versus Anastrozole in Postmenopausal Patients With Hormone Receptor-Positive, Node-Positive Early Breast Cancer: Final Results of the Randomized Phase III Femara Versus Anastrozole Clinical Evaluation (FACE) Trial. J Clin Oncol 2017 35:1041-1048. PMID 28113032
13. Colleoni M, et al. Extended adjuvant intermittent letrozole versus continuous letrozole in postmenopausal women with breast cancer (SOLE): a multicentre, open-label, randomised, phase 3 trial. Lancet Oncol 2017 19:127-138. PMID 29158011
14. Goss PE, Ingle JN, Pritchard KI, et al. Extending aromatase-inhibitor adjuvant therapy to 10 years. N Engl J Med 2016 375:209-219. PMID 27264120
15. Davies C, Pan H, Godwin J, et al. Long-term effects of continuing adjuvant tamoxifen to 10 years versus stopping at 5 years after diagnosis of oestrogen receptor-positive breast cancer: ATLAS, a randomised trial. Lancet 2013 381:805-816. PMID 23219286
16. Goss PE, et al. Exemestane versus anastrozole in postmenopausal women with early breast cancer: NCIC CTG MA.27--a randomized controlled phase III trial. J Clin Oncol 2013 31:1398-404. PMID 23358971
17. Regan MM, Neven P, Giobbie-Hurder A, et al. Assessment of letrozole and tamoxifen alone and in sequence for postmenopausal women with steroid hormone receptor-positive breast cancer: the BIG 1-98 randomised clinical trial at 8.1 years median follow-up. Lancet Oncol 2011 12:1101-1108. PMID 22018631
18. van de Velde CJH, Rea D, Seynaeve C, et al. Adjuvant tamoxifen and exemestane in early breast cancer (TEAM): a randomised phase 3 trial. Lancet 2011 377:321-331. PMID 21247627
19. Cuzick J, Sestak I, Baum M, et al. Effect of anastrozole and tamoxifen as adjuvant treatment for early-stage breast cancer: 10-year analysis of the ATAC trial. Lancet Oncol 2010 11:1135-1141. PMID 21087898
20. Gnant M, Mlineritsch B, Schippinger W, et al. Endocrine therapy plus zoledronic acid in premenopausal breast cancer. N Engl J Med 2009 360:679-691. PMID 19213681
21. Albain KS, et al. Adjuvant chemotherapy and timing of tamoxifen in postmenopausal patients with endocrine-responsive, node-positive breast cancer: a phase 3, open-label, randomised controlled trial. Lancet 2009 374:2055-2063. PMID 20004966
22. Coombes RC, Kilburn LS, Snowdon CF, et al. Survival and safety of exemestane versus tamoxifen after 2-3 years' tamoxifen treatment (Intergroup Exemestane Study): a randomised controlled trial. Lancet 2007 369:559-570. PMID 17307102
23. Goss PE, Ingle JN, Martino S, et al. A randomized trial of letrozole in postmenopausal women after five years of tamoxifen therapy for early-stage breast cancer. N Engl J Med 2003 349:1793-1802. PMID 14551341
24. Fisher B, et al. Tamoxifen and chemotherapy for lymph node-negative, estrogen receptor-positive breast cancer. J Natl Cancer Inst 1997 89:1673-82. PMID 9390536
25. Fisher B, et al. A randomized clinical trial evaluating tamoxifen in the treatment of patients with node-negative breast cancer who have estrogen-receptor-positive tumors. N Engl J Med 1989 320:479-84. PMID 2644532
26. Jakesz R, et al. Switching of postmenopausal women with endocrine-responsive early breast cancer to anastrozole after 2 years' adjuvant tamoxifen: combined results of ABCSG trial 8 and ARNO 95 trial. Lancet 366:455-62. PMID 16084253
27. Hughes KS, Schnaper LA, Bellon JR, et al. Lumpectomy plus tamoxifen with or without irradiation in women age 70 years or older with early breast cancer: long-term follow-up of CALGB 9343. J Clin Oncol 2013 31:2382-2387. PMID 23690420
28. Partridge AH, Niman SM, Ruggeri M, et al. Interrupting Endocrine Therapy to Attempt Pregnancy after Breast Cancer. N Engl J Med 2023 388:1645-1656. PMID 37133584
29. Bartlett JMS, et al. Breast Cancer Index Is a Predictive Biomarker of Treatment Benefit and Outcome from Extended Tamoxifen Therapy: Final Analysis of the Trans-aTTom Study. Clin Cancer Res 2022 28:1871-1880. PMID 35144966
30. Hamilton EP, Loibl S, Bachelot T, et al. CAMBRIA-1 & CAMBRIA-2 phase III trials: camizestrant versus standard endocrine therapy in ER+/HER2- early breast cancer. Future Oncol 2025 21:795-806. PMID 40017004
31. Paik S, et al. A multigene assay to predict recurrence of tamoxifen-treated, node-negative breast cancer. N Engl J Med 2004 351:2817-26. PMID 15591335
32. Albain KS, et al. Prognostic and predictive value of the 21-gene recurrence score assay in postmenopausal women with node-positive, oestrogen-receptor-positive breast cancer on chemotherapy: a retrospective analysis of a randomised trial. Lancet Oncol 2009 11:55-65. PMID 20005174
33. Pan H, Gray R, Braybrooke J, et al. 20-year risks of breast-cancer recurrence after stopping endocrine therapy at 5 years. N Engl J Med 2017 377:1836-1846. PMID 29117498
34. Early Breast Cancer Trialists' Collaborative Group (EBCTCG). Relevance of breast cancer hormone receptors and other factors to the efficacy of adjuvant tamoxifen: patient-level meta-analysis of randomised trials. Lancet 2011 378:771-784. PMID 21802721
35. Early Breast Cancer Trialists' Collaborative Group (EBCTCG). Tamoxifen for early breast cancer: an overview of the randomised trials. Early Breast Cancer Trialists' Collaborative Group. Lancet 1998 351:1451-67. PMID 9605801
36. Early Breast Cancer Trialists' Collaborative Group (EBCTCG). Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomised trials. Lancet 2005 365:1687-717. PMID 15894097
37. Pegram MD, Pauletti G, Slamon DJ. HER-2/neu as a predictive marker of response to breast cancer therapy. Breast Cancer Res Treat 1998 52:65-77. PMID 10066073
38. Piccart MJ, Di Leo A, Hamilton A. HER2. a 'predictive factor' ready to use in the daily management of breast cancer patients?. Eur J Cancer 2000 36:1755-61. PMID 10974622
39. Mass R. The role of HER-2 expression in predicting response to therapy in breast cancer. Semin Oncol 2000 27:46-52; discussion 92-100. PMID 11236028
40. Berry DA, Muss HB, Thor AD, et al. HER-2/neu and p53 expression versus tamoxifen resistance in estrogen receptor-positive, node-positive breast cancer. J Clin Oncol 2000 18:3471-9. PMID 11032587
41. Knoop AS, Bentzen SM, Nielsen MM, Rasmussen BB, Rose C. Value of epidermal growth factor receptor, HER2, p53, and steroid receptors in predicting the efficacy of tamoxifen in high-risk postmenopausal breast cancer patients. J Clin Oncol 2001 19:3376-84. PMID 11454885
42. De Laurentiis M, Arpino G, Massarelli E, et al. A meta-analysis on the interaction between HER-2 expression and response to endocrine treatment in advanced breast cancer. Clin Cancer Res 2005 11:4741-8. PMID 16000569
43. Arpino G, Green SJ, Allred DC, et al. HER-2 amplification, HER-1 expression, and tamoxifen response in estrogen receptor-positive metastatic breast cancer: a southwest oncology group study. Clin Cancer Res 2004 10:5670-6. PMID 15355892
44. Eppenberger-Castori S, Kueng W, Benz C, et al. Prognostic and predictive significance of ErbB-2 breast tumor levels measured by enzyme immunoassay. J Clin Oncol 2001 19:645-56. PMID 11157014
45. Dowsett M, Allred C, Knox J, et al. Relationship between quantitative estrogen and progesterone receptor expression and human epidermal growth factor receptor 2 (HER-2) status with recurrence in the Arimidex, Tamoxifen, Alone or in Combination trial. J Clin Oncol 2008 26:1059-65. PMID 18227529
46. Early Breast Cancer Trialists' Collaborative Group (EBCTCG). Aromatase inhibitors versus tamoxifen in early breast cancer: patient-level meta-analysis of the randomised trials. Lancet 2015 386:1341-1352. PMID 26211827
47. Early Breast Cancer Trialists' Collaborative Group (EBCTCG). Aromatase inhibitors versus tamoxifen in premenopausal women with oestrogen receptor-positive early-stage breast cancer treated with ovarian suppression: a patient-level meta-analysis of 7030 women from four randomised trials. Lancet Oncol 2022 23:382-392. PMID 35123662
48. Baum M, Budzar AU, Cuzick J, et al. Anastrozole alone or in combination with tamoxifen versus tamoxifen alone for adjuvant treatment of postmenopausal women with early breast cancer: first results of the ATAC randomised trial. Lancet 2002 359:2131-9. PMID 12090977
49. Howell A, et al. Results of the ATAC (Arimidex, Tamoxifen, Alone or in Combination) trial after completion of 5 years' adjuvant treatment for breast cancer. Lancet 2005 365:60-2. PMID 15639680
50. Forbes JF, et al. Effect of anastrozole and tamoxifen as adjuvant treatment for early-stage breast cancer: 100-month analysis of the ATAC trial. Lancet Oncol 2008 9:45-53. PMID 18083636
51. Thürlimann B, et al. A comparison of letrozole and tamoxifen in postmenopausal women with early breast cancer. N Engl J Med 2005 353:2747-57. PMID 16382061
52. BIG 1-98 Collaborative Group, Mouridsen H, Giobbie-Hurder A, et al. Letrozole therapy alone or in sequence with tamoxifen in women with breast cancer. N Engl J Med 2009 361:766-76. PMID 19692688
53. Coombes RC, et al. A randomized trial of exemestane after two to three years of tamoxifen therapy in postmenopausal women with primary breast cancer. N Engl J Med 2004 350:1081-92. PMID 15014181
54. Kaufmann M, Jonat W, Hilfrich J, et al. Improved overall survival in postmenopausal women with early breast cancer after anastrozole initiated after treatment with tamoxifen compared with continued tamoxifen: the ARNO 95 Study. J Clin Oncol 2007 25:2664-70. PMID 17563395
55. Boccardo F, Rubagotti A, Puntoni M, et al. Switching to anastrozole versus continued tamoxifen treatment of early breast cancer: preliminary results of the Italian Tamoxifen Anastrozole Trial. J Clin Oncol 2005 23:5138-47. PMID 16009955
56. Boccardo F, Rubagotti A, Guglielmini P, et al. Switching to anastrozole versus continued tamoxifen treatment of early breast cancer. Updated results of the Italian tamoxifen anastrozole (ITA) trial. Ann Oncol 2006 17 Suppl 7:vii10-4. PMID 16760270
57. Jonat W, Gnant M, Boccardo F, et al. Effectiveness of switching from adjuvant tamoxifen to anastrozole in postmenopausal women with hormone-sensitive early-stage breast cancer: a meta-analysis. Lancet Oncol 2006 7:991-6. PMID 17138220
58. Dowsett M, Cuzick J, Ingle J, et al. Meta-analysis of breast cancer outcomes in adjuvant trials of aromatase inhibitors versus tamoxifen. J Clin Oncol 2010 28:509-18. PMID 19949017
59. Eastell R, Adams JE, Coleman RE, et al. Effect of anastrozole on bone mineral density: 5-year results from the anastrozole, tamoxifen, alone or in combination trial 18233230. J Clin Oncol 2008 26:1051-7. PMID 18309940
60. Duffy S, Jackson TL, Lansdown M, et al. The ATAC ('Arimidex', Tamoxifen, Alone or in Combination) adjuvant breast cancer trial: first results of the endometrial sub-protocol following 2 years of treatment. Hum Reprod 2006 21:545-53. PMID 16210385
61. Fallowfield L, Cella D, Cuzick J, Francis S, Locker G, Howell A. Quality of life of postmenopausal women in the Arimidex, Tamoxifen, Alone or in Combination (ATAC) Adjuvant Breast Cancer Trial. J Clin Oncol 2004 22:4261-71. PMID 15514369
62. Buzdar AU, Guastalla JP, Nabholtz JM, Cuzick J, ATAC Trialists' Group. Impact of chemotherapy regimens prior to endocrine therapy: Results from the ATAC (Anastrozole and Tamoxifen, Alone or in Combination) trial. Cancer 2006 107:472-80. PMID 16804925
63. Dowsett M, Cuzick J, Howell A, Jackson I. Pharmacokinetics of anastrozole and tamoxifen alone, and in combination, during adjuvant endocrine therapy for early breast cancer in postmenopausal women: a sub-protocol of the 'Arimidex and tamoxifen alone or in combination' (ATAC) trial. Br J Cancer 2001 85:317-24. PMID 11487258
64. Rabaglio M, Sun Z, Price KN, et al. Bone fractures among postmenopausal patients with endocrine-responsive early breast cancer treated with 5 years of letrozole or tamoxifen in the BIG 1-98 trial. Ann Oncol 2009 20:1489-1498. PMID 19474112
65. Mouridsen H, Keshaviah A, Coates AS, et al. Cardiovascular adverse events during adjuvant endocrine therapy for early breast cancer using letrozole or tamoxifen: safety analysis of BIG 1-98 trial. J Clin Oncol 2007 25:5715-22. PMID 17998546
66. Smith IE, Dowsett M, Yap YS, et al. Adjuvant aromatase inhibitors for early breast cancer after chemotherapy-induced amenorrhoea: caution and suggested guidelines. J Clin Oncol 2006 24:2444-7. PMID 16735701
67. Yu B, Douglas N, Ferin MJ, et al. Changes in markers of ovarian reserve and endocrine function in young women with breast cancer undergoing adjuvant chemotherapy. Cancer 2010 116:2099-105. PMID 20187091
68. Francis PA, Regan MM, Fleming GF, et al. Adjuvant ovarian suppression in premenopausal breast cancer. N Engl J Med 2015 372:436-446. PMID 25495490
69. Francis PA, et al. Adjuvant Endocrine Therapy in Premenopausal Breast Cancer: 12-Year Results From SOFT. J Clin Oncol 2022 41:1370-1375. PMID 36493334
70. Baek SY, et al. Adding Ovarian Suppression to Tamoxifen for Premenopausal Women With Hormone Receptor-Positive Breast Cancer After Chemotherapy: An 8-Year Follow-Up of the ASTRRA Trial. J Clin Oncol 2023 41:4864-4871. PMID 37607321
71. Pagani O, Walley BA, Fleming GF, et al. Adjuvant exemestane with ovarian suppression in premenopausal breast cancer: long-term follow-up of the combined TEXT and SOFT trials. J Clin Oncol 2023 41:1376-1382. PMID 36521078
72. Gnant M, et al. Zoledronic acid combined with adjuvant endocrine therapy of tamoxifen versus anastrozol plus ovarian function suppression in premenopausal early breast cancer: final analysis of the Austrian Breast and Colorectal Cancer Study Group Trial 12. Ann Oncol 2014 26:313-20. PMID 25403582
73. Mamounas EP, et al. Ten-year update: NRG Oncology/National Surgical Adjuvant Breast and Bowel Project B-42 randomized trial: extended letrozole therapy in early-stage breast cancer. J Natl Cancer Inst 2023 115:1302-1309. PMID 37184928
74. Jerusalem G, et al. Continuous versus intermittent extended adjuvant letrozole for breast cancer: final results of randomized phase III SOLE (Study of Letrozole Extension) and SOLE Estrogen Substudy. Ann Oncol 2021 32:1256-1266. PMID 34384882
75. Ingle JN, et al. Intent-to-treat analysis of the placebo-controlled trial of letrozole for extended adjuvant therapy in early breast cancer: NCIC CTG MA.17. Ann Oncol 2008 19:877-82. PMID 18332043
76. Goss PE, et al. Randomized trial of letrozole following tamoxifen as extended adjuvant therapy in receptor-positive breast cancer: updated findings from NCIC CTG MA.17. J Natl Cancer Inst 2005 97:1262-71. PMID 16145047
77. Jin H, Tu D, Zhao N, Shepherd LE, Goss PE. Longer-term outcomes of letrozole versus placebo after 5 years of tamoxifen in the NCIC CTG MA.17 trial: analyses adjusting for treatment crossover. J Clin Oncol 2012 30:718-21. PMID 22042967
78. Goss PE, Ingle JN, Pater JL, et al. Late extended adjuvant treatment with letrozole improves outcome in women with early-stage breast cancer who complete 5 years of tamoxifen. J Clin Oncol 2008 26:1948-55. PMID 18332475
79. Perez EA, Josse RG, Pritchard KI, et al. Effect of letrozole versus placebo on bone mineral density in women with primary breast cancer completing 5 or more years of adjuvant tamoxifen: a companion study to NCIC CTG MA.17. J Clin Oncol 2006 24:3629-35. PMID 16822845
80. Whelan TJ, Goss PE, Ingle JN, et al. Assessment of quality of life in MA.17: a randomized, placebo-controlled trial of letrozole after 5 years of tamoxifen in postmenopausal women. J Clin Oncol 2005 23:6931-40. PMID 16157934
81. Jakesz R, Greil R, Gnant M, et al. Extended adjuvant therapy with anastrozole among postmenopausal breast cancer patients: results from the randomized Austrian Breast and Colorectal Cancer Study Group Trial 6a. J Natl Cancer Inst 2007 99:1845-53. PMID 18073378
82. Early Breast Cancer Trialists' Collaborative Group (EBCTCG). Extending the duration of endocrine treatment for early breast cancer: patient-level meta-analysis of 12 randomised trials of aromatase inhibitors in 22 031 postmenopausal women already treated with at least 5 years of endocrine therapy. Lancet 2025 406:603-614. PMID 40783288
83. Colleoni M, Sun Z, Price KN, et al. Annual hazard rates of recurrence for breast cancer during 24 years of follow-up: results from the International Breast Cancer Study Group trials I to V. J Clin Oncol 2016 34:927-935. PMID 26786933
84. Saphner T, Tormey DC, Gray R. Annual hazard rates of recurrence for breast cancer after primary therapy. J Clin Oncol 1996 14:2738-2746. PMID 8874335
85. Hershman DL, Kushi LH, Shao T, et al. Early discontinuation and nonadherence to adjuvant hormonal therapy in a cohort of 8,769 early-stage breast cancer patients. J Clin Oncol 2010 28:4120-4128. PMID 20585090
86. Lazzeroni M, Puntoni M, Guerrieri-Gonzaga A, et al. Randomized placebo controlled trial of low-dose tamoxifen to prevent recurrence in breast noninvasive neoplasia: a 10-year follow-up of TAM-01 study. J Clin Oncol 2023 41:3116-3121. PMID 36917758
87. Hamann M, Christgen M, Gluz O, et al. Endocrine response assessment in HR-positive HER2-negative early breast cancer: concordance of local versus central Ki67 measurements in the WSG ADAPTcycle trial (n = 5292). ESMO Open 2025 10:105552. PMID 40815984
88. Fjermeros K, Ghannoum S, Geisler SB, et al. The NEOLETRIB trial: neoadjuvant treatment with letrozole and ribociclib in ER-positive, HER2-negative breast cancer. Future Oncol 2024 20:2457-2466. PMID 39073142
89. Ma XJ, Wang Z, Ryan PD, et al. A two-gene expression ratio predicts clinical outcome in breast cancer patients treated with tamoxifen. Cancer Cell 2004 5:607-16. PMID 15193263
90. Sgroi DC, Sestak I, Cuzick J, et al. Prediction of late distant recurrence in patients with oestrogen-receptor-positive breast cancer: a prospective comparison of the breast-cancer index (BCI) assay, 21-gene recurrence score, and IHC4 in the TransATAC study population. Lancet Oncol 2013 14:1067-1076. PMID 24035531
91. Noordhoek I, Treuner K, Putter H, et al. Breast Cancer Index Predicts Extended Endocrine Benefit to Individualize Selection of Patients with HREarly-stage Breast Cancer for 10 Years of Endocrine Therapy. Clin Cancer Res 2021 27:311-319. PMID 33109739
92. Bartlett JMS, Sgroi DC, Treuner K, et al. Breast Cancer Index and prediction of benefit from extended endocrine therapy in breast cancer patients treated in the Adjuvant Tamoxifen-To Offer More? (aTTom) trial. Ann Oncol 2019 30:1776-1783. PMID 31504126