Management of Early and Metastatic Breast Cancer: A Consensus Statement from Oncologists in Saudi Arabia.

Key Summary Points

Introduction
Breast cancer (BC) is the most common cancer diagnosis and leading cancer-related cause of mortality among women globally [1]. Excluding nonmelanoma skin cancer, one-quarter of all new cancer cases among women are BC, contributing to 15.5% of cancer-related deaths [1]. The prevalence of BC and its associated mortality rate varies widely by region and country due to differences in population age demographics, risk factor prevalence, and healthcare resource availability [1].
In the Kingdom of Saudi Arabia (hereafter referred to as Saudi Arabia), BC is the most common cancer among women [2], representing approximately 30% of all cancer cases [3, 4]. Among both sexes, the incidence rate is estimated to be 14.8% and the mortality rate 8.5% [3]. Of concern is the increase in BC incidence among Saudi women between 2001 and 2017 [4], reflected in a statistically significant increase in BC prevalence in Saudi Arabia between 2010 and 2018 [3]. The Ministry of Health in Saudi Arabia states that more than one-half of all BC cases are detected at a late stage, leading to a higher mortality rate and higher treatment costs [2]. Early detection is recognized by the Saudi government as essential for treatment success and for reducing mortality rates [2]. These goals are in line with the Global Breast Cancer Initiative (GBCI) launched by the World Health Organization in 2021, where one of the three key aims is to make > 60% of diagnoses at stage I or II disease [1].
The approach to treatment of BC depends on whether the disease is diagnosed at an early stage or as metastatic disease [5, 6]. While clinical practice guidelines are available for Saudi oncologists from the Saudi Health Council’s National Cancer Center, these were last published in 2020 [7]. Subsequently, there have been changes in types of treatment modalities available (cell-based therapies, gene therapy, antibody–drug conjugates [ADCs]), and the publication of phase 2/3 clinical trial and long-term follow-up data for novel drugs. In our view, current clinical practice in Saudi Arabia is generally in line with international recommendations, but these guidelines may recommend agents that are not easily available in Saudi Arabia or that are of significant cost burden to the Saudi state. In Saudi Arabia, cancer treatment is provided free of charge through governmental institutions, which ensures patient access but also places a substantial financial burden on the national healthcare system. The escalating costs of novel oncology therapeutics (such as immune checkpoint inhibitors [ICIs] and ADCs) necessitate careful consideration of cost-effectiveness and sustainability. Each hospital, whether under the Ministry of Health or operating as an independent tertiary center, typically functions within a limited annual budget for innovative cancer drugs. Consequently, oncologists often face complex decisions regarding patient prioritization for high-cost therapies, constricted by the aim to allocate resources to those most likely to achieve meaningful clinical benefit. These real-world constraints contribute to variability in treatment approaches and underscore the importance of developing a unified, evidence-based consensus to support fair and rational decision-making across institutions.
There is, therefore, an unmet need for local guidance on the management of BC for the general medical oncologist in Saudi Arabia, taking into consideration advances in treatment options, treatment availability, and cost (financial ‘toxicity’). To address this need, we convened a panel of expert clinical oncologists to harness their collective expertise and develop consensus statements that specifically address the challenges in BC care in Saudi Arabia, where robust evidence is lacking. These statements aim to provide medical oncologists in Saudi Arabia with practical guidance for the management of controversial or uncertain clinical scenarios, thereby unifying clinical practice for improved patient care and providing guidance in areas where international guidelines may not fully address local needs or resources.

Methods
A modified Delphi process was used to develop a set of consensus statements. Experts from clinical institutions across Saudi Arabia were invited to participate in the panel by S. Alaklabi from King Faisal Specialist Hospital & Research Centre (KFSHRC; Riyadh, Saudi Arabia). These panelists were identified based on their professional qualifications, clinical experience in the treatment of BC, professional affiliations, leadership roles, research and academic contributions, and experience of working in multidisciplinary teams. All panelists had to be affiliated with a recognized cancer treatment center or academic medical center in Saudi Arabia and to disclose their potential conflicts of interest. The selected panelists ensured representation from different regions of Saudi Arabia and different practice settings (i.e., academic, tertiary, or rural). The core group of panelists comprised 23 individuals (the authors) who participated in both rounds of voting.
The process of consensus statement development included a pragmatic literature review (see the literature search and data synthesis in the Electronic Supplementary Material file), an initial on-line survey in which panelists (n = 26) voted anonymously on statements drafted by S. Alaklabi and M. Alzahrani (Round 1 of voting), and a subsequent satellite meeting held on 1 February 2025 at the inaugural KFSHRC Breast Cancer Conference in Riyadh. At this meeting, the results of the survey for each statement were presented to the panel and meeting audience (KFSHRC Breast Cancer Conference attendees). These results were discussed to ensure: (1) transparency in the decision-making process; (2) broader engagement with the meeting audience; and (3) in-depth discussion of areas of disagreement. Based on these discussions, statements could be reformulated if necessary, and then voted on by panelists in real-time (Round 2 of voting), with the results shared with the meeting audience. This approach was used to enhance engagement, provide immediate feedback, and allow for real-time transparent consensus building.
There were 35 consensus statements framed as questions, with 20 on early BC and 15 on metastatic BC. Each question could be answered ‘Yes,’ ‘No,’ or ‘Abstain.’ As is typical for Delphi consensus statements, the threshold for consensus was designated as ≥ 70% of votes for either ‘Yes’ or ‘No’. ‘Abstain’ votes were included in the denominator when calculating proportion of ‘Yes’ and ‘No’ votes.

Ethics/Ethical Approval
As this was a Delphi-based consensus document, institutional review board approval was not required. However, all participants in the study were aware of the objectives of the study and that the results of the study would be published. Before the start of the consensus procedure, each participant was briefed about the nature of the activity. All participants have given their independent opinion based on their previous experience. The final responses of all participants were recorded anonymously to ensure the protection of their individual data.

Results
The core expert panel comprised medical oncologists from five locations across Saudi Arabia (Riyadh, Jeddah, Dammam, Makkah, and Jazan). The first round of voting was completed online by 26 panelists (100%) approximately 7 days before the second round of voting, which was completed in person by 19 to 23 panelists (during the conference; not all panelists voted on every statement).
Voting results are presented in Tables 1 and 2. Of the 20 statements on early BC, seven achieved consensus in the first round (Table 1). The panel agreed to omit statement 14 (S14) on early BC from the second round of voting. After reformulation of eight statements, consensus was achieved for 13 statements in the second round of voting. Of the 15 statements on metastatic BC, four achieved consensus in the first round, and an additional three statements achieved consensus in the second round (Table 2). Recommendations for the management of early and metastatic BC based on the statements for which consensus was reached are outlined in Table 3.

Discussion

Early BC

Hormone Receptor-Positive BC
The focus for patients with hormone receptor-positive (HR+) early BC was to obtain consensus on statements relating to neoadjuvant and adjuvant systemic targeted therapy (S1–S4; Table 1). There were no statements on local therapies (surgery, radiotherapy [RT]). Two cyclin-dependent kinase 4/6 (CDK4/6) inhibitors are options for adjuvant therapy: abemaciclib and ribociclib. The efficacy and tolerability of abemaciclib was investigated in the monarchE trial [8], and those of ribociclib were investigated in the NATALEE trial [9], both randomized phase 3 trials in patients with HR+, human epidermal growth factor receptor 2 (HER2)-negative (HER2−; also known as ERBB2−) early BC. The monarchE trial enrolled patients with high-risk, N+ early BC, defined by specific clinical and pathological criteria (rather than by Oncotype DX recurrence score categories). Among the panelists, there was consensus that neither ribociclib or abemaciclib should be considered the preferred adjuvant CDK4/6 inhibitor; for patients eligible for either agent, the panelists agreed that both ribociclib and abemaciclib are acceptable options, administered according to the NATALEE and monarchE treatment protocols, respectively. However, some panelists were in favor of abemaciclib due to the availability of longer follow-up data. Consensus was not reached on whether CDK4/6 inhibitor adjuvant therapy may be omitted in patients with a low Oncotype DX recurrence score (i.e., < 26 [S4]). While some panelists considered the omission of adjuvant ribociclib reasonable for patients with T2 M0 N0 disease and a low Oncotype DX recurrence score, other panelists noted that the findings of the monarchE trial demonstrated benefit with abemaciclib irrespective of the Oncotype DX recurrence score [8].
In general, adjuvant endocrine therapy is used to treat patients with invasive HR+ BC of any stage and HER2 status [5]. The panelists confirmed their agreement—that in this setting (S5) adjuvant endocrine therapy should be prescribed even in patients with low estrogen receptor (ER) expression (< 10%), in addition to other indicated systemic therapeutic modalities. The European Society of Medical Oncology (ESMO) guidelines describe patients with HER2− BC and low (1–9%) ER expression and/or low progesterone receptor (PgR) expression as a heterogeneous group of patients who behave biologically similarly to patients with triple-negative BC (TNBC) [5]. ESMO guidelines advise that this patient population might have a better response to chemotherapy than to endocrine therapy.
To clarify the use of endocrine therapy in HR+ BC, the uncertainty of the ideal treatment duration in postmenopausal women in the neoadjuvant setting was addressed. The consensus was that the ideal minimum treatment duration is 6 months and the ideal maximum treatment is 12 months (S6, S7) [10]. An observational study reported an objective response rate with neoadjuvant letrozole of around 50% at month 4, 85% at month 8, and 95% at month 12 in patients with ER+ BC [10]. However, the majority of the clinical evidence supports an optimal treatment duration of < 12 months. Data from a 12-month phase 2 trial (CARABELA) found that neoadjuvant endocrine therapy + abemaciclib was not as efficacious as standard chemotherapy for HR+/HER2− BC [11]. In the panelists’ experience, the duration of neoadjuvant endocrine therapy is typically 4–6 months. Factors to consider in determining the duration of neoadjuvant treatment include the goal of neoadjuvant treatment, whether the patient is a candidate for surgery, and the risk of disease progression. While a meta-analysis found a lower mastectomy rate in patients undergoing neoadjuvant endocrine therapy versus surgery followed by adjuvant endocrine therapy (risk ratio [RR] 0.53; 95% confidence interval [CI] 0.44–0.64), there was no significant difference in mastectomy rate by duration of neoadjuvant endocrine therapy (> 4 vs ≤ 4 months) [12]. Monitoring disease progression using both clinical and imaging assessment is considered critical during longer-term neoadjuvant endocrine therapy.

Triple-Negative BC
According to ESMO guidelines, initial treatment approaches for early-stage TNBC are either surgery with or without whole breast RT for individuals with cT1a/b N0 status or neoadjuvant chemotherapy or chemoimmunotherapy before surgery (± RT) for individuals with either cT1c–T4 or N+ status [5].

Patients With pCR to Neoadjuvant Chemotherapy
Panelists were asked whether adjuvant pembrolizumab treatment can be omitted (i.e., not continued) when a patient with TNBC has pathologic complete response (pCR) to neoadjuvant chemoimmunotherapy (S13). Consensus was not reached for this statement because panelists suggested that initial tumor size and type of neoadjuvant regimen needs to be considered when deciding whether or not to offer adjuvant pembrolizumab in this clinical scenario. In the KEYNOTE-522 trial, the survival curves in patients with a pCR were similar between treatment groups [13]; however, this observation does not definitively demonstrate that pembrolizumab can be safely discontinued in these patients, and the panelists recommend waiting for longer-term data. A large phase 3 trial (OPTIMICE-pCR; NCT05812807) is currently ongoing to determine whether adjuvant pembrolizumab can be safely omitted in those patients who achieve pCR to chemoimmunotherapy by randomizing patients to pembrolizumab for an additional 9 cycles versus surveillance in the adjuvant setting. Pertaining to oncologic practice in Saudi Arabia, panelists highlighted concerns relating to pembrolizumab toxicity and the financial toxicity associated with pembrolizumab treatment for no significant gain in overall survival (OS).

Patients Without pCR to Neoadjuvant Chemotherapy
The panelists recommended that patients with any amount of residual disease post neoadjuvant therapy are eligible for adjuvant therapy, whether with chemotherapy (capecitabine), the ICI pembrolizumab, or the poly ADP-ribose polymerase (PARP) inhibitor olaparib (S17). Consensus was sought on several questions relating to systemic adjuvant therapy in patients with TNBC who do not achieve pCR after neoadjuvant chemoimmunotherapy (S8–S12). Statement 8 pertains to patients with BReast CAncer gene (BRCA)-mutant TNBC, and S9–S12 pertain to patients with BRCA wild-type status.
In patients with BRCA-mutant TNBC who have residual disease after neoadjuvant chemoimmunotherapy (containing olaparib or when olaparib is planned and available for the patient), the consensus was to avoid use of adjuvant capecitabine (i.e., capecitabine after olaparib [S8]). Panelists raised concerns regarding cumulative toxicity and the conflicting evidence on the risk of myelodysplastic syndrome and leukemia associated with PARP inhibitors [14]. PARP inhibitors are sometimes associated with significant and serious organ toxicity, especially hematologic [15] and several GI-related toxicities (nausea, vomiting) [16]. Additionally, the exact benefit of capecitabine in patients with BRCA-mutant BC is not clear, nor are there sufficient data on the safety and efficacy of capecitabine after 1 year of olaparib treatment.
The panelists agreed that clinicians can consider adding capecitabine to pembrolizumab (in patients continuing pembrolizumab [S9]), in agreement with ESMO recommendations [5], although there is no definitive evidence to support this approach and it is possible that patient outcomes remain poor. Further, the panelists recommended continuing pembrolizumab in combination with capecitabine as adjuvant therapy, rather than discontinuing pembrolizumab and prescribing capecitabine alone as adjuvant therapy (S10), or prescribe pembrolizumab alone (without capecitabine). The panelists mentioned one clinical scenario in which immunotherapy could be discontinued, namely when patients experience disease progression during neoadjuvant therapy with pembrolizumab (or other immunotherapy) + chemotherapy. Currently, phase 3 trial evidence for adjuvant capecitabine alone for patients with TNBC who have residual disease after neoadjuvant therapy is limited to patients who received neoadjuvant chemotherapy (no immunotherapy) in the CREATE-X trial [17]. In this trial, the subgroup of patients with TNBC and residual invasive disease experienced good outcomes with adjuvant capecitabine after neoadjuvant chemotherapy, with a hazard ratio (HR) of 0.58 for recurrence, second cancer, or death, and of 0.52 for OS [17]. There was a significant disease-free survival (DFS) advantage in the TNBC subgroup receiving adjuvant capecitabine (5-year DFS rate of 69.8%) [17]. Real-world evidence supports this treatment approach, with adjuvant capecitabine significantly improving both distant-free DFS and OS rates (86.3% and 93.3%, respectively) in a Breast Cancer Registry involving Korean patients with TNBC and residual disease, compared with corresponding rates for those who did not receive any adjuvant treatment (74.4% and 83.8%, respectively) [18].
The panelists also agreed that adjuvant capecitabine is to be avoided after completion of adjuvant pembrolizumab (S12). However, sequential therapy may be acceptable when there is proven toxicity with combination pembrolizumab and capecitabine. Consensus could not be reached on the question of continuing pembrolizumab (from the neoadjuvant through to the adjuvant setting) without adding capecitabine (S11). It was highlighted that although data are limited, any DFS advantage from adding capecitabine may not be justified by the additional toxicity. Yet patients with TNBC are at the highest risk of recurrence and death within 5 years of diagnosis compared with those with other BC subtypes [19], further complicating the decision regarding continuing pembrolizumab with or without capecitabine. Notably, the majority response (67% of panelists voted ‘no’) for S11 was consistent with response to S10 (do not discontinue pembrolizumab and give capecitabine alone as adjuvant treatment). Nevertheless, the lack of consensus on S11 is unsurprising given that data on the use of pembrolizumab (without capecitabine) in the adjuvant setting are currently available only from one trial in TNBC (KEYNOTE-522 [20]). The 5-year event-free survival (EFS) rate in KEYNOTE-522 was 62.6% with pembrolizumab [20]. In the panelists’ experience, most clinicians would prescribe pembrolizumab (without capecitabine) as adjuvant treatment using the pembrolizumab regimen described in the KEYNOTE-522 trial [13, 21]. In considering the choice of adjuvant therapy for patients with BRCA wild-type TNBC who have residual disease after neoadjuvant chemoimmunotherapy, the panelists discussed whether the level of residual disease (Residual Cancer Burden [RCB]-1, -2, or -3) was relevant. RCB is prognostic of EFS, with higher RCB scores significantly associated with worse outcomes [22]. The combination of adjuvant pembrolizumab + capecitabine can be considered in patients with RCB-3 or possibly RCB-2.
While discussion was focused on pembrolizumab and/or capecitabine for patients with BRCA wild-type BC, adjuvant olaparib was mentioned as a rational choice for patients with BRCA-mutant BC and RCB-3. This is consistent with ESMO guidance for patients with BRCA1/2-mutant and high-risk TNBC (pCR not achieved after neoadjuvant treatment or pathologic stage II–III) [5].
Regarding the choice of adjuvant chemotherapy in patients with TNBC, the panelists recommended that patients with T1b N0 disease receive standard chemotherapy with 6 cycles of carboplatin/paclitaxel weekly or 4 cycles of docetaxel/cyclophosphamide (S16), whereas adjuvant chemotherapy can be omitted in patients with T1a N0 disease (S15).

Human Epidermal Growth Factor Receptor 2-Positive BC
Based on the results of the KATHERINE trial, adjuvant trastuzumab emtansine (T-DM1) became the standard of care for patients with HER2+ early BC with residual invasive disease (breast and/or axillary nodes) at the time of surgery who had received taxane- and trastuzumab-containing neoadjuvant therapy (± anthracyclines) [23]. The panelists sought to clarify the appropriate patient population for treatment escalation to T-DM1, but consensus was not reached on whether T-DM1 should be limited to patients with residual disease > 5 mm in size. International guidelines vary in their recommendations for adjuvant therapy in cases of residual disease in this patient population, with the ESMO recommending T-DM1 without specifying criteria for residual disease size [5]. While some panelists emphasized the benefit of T-DM1 in patients with residual invasive disease, including those with microinvasive disease in the KATHERINE trial (ypT0, ypT1a, ypT1b, YpT1mic tumors at surgery; n = 331) [23], others expressed doubt about the level of statistical significance given the wide 95% CIs for this subgroup (HR for invasive-disease event 0.66; 95% CI 0.44–1.00) [23]. However, subgroup analyses from KATHERINE showed that the efficacy of T-DM1 was consistent in all subgroups (i.e., regardless of residual disease extent at surgery [invasive disease vs microinvasion]), HR status, single or dual HER2-targeted neoadjuvant therapy, and patient baseline characteristics [23].
Given the overall very favorable results from KATHERINE for T-DM1 versus trastuzumab (invasive DFS: HR 0.54, 95% CI 0.44–0.66; OS HR: 0.66, 95% CI 0.51–0.87; p = 0.003) [24], some panelists stated that they would prescribe T-DM1 to patients with microinvasive residual cancer.

Multiple Subtypes
In patients with early BC at high risk of recurrence, the panelists discussed sequential targeted therapy and combination targeted therapy options. Consensus was not quite reached on the use of adjuvant CDK4/6 inhibitors (palbociclib, ribociclib, or abemaciclib) sequentially after olaparib in patients with BRCA-mutant BC (67% of panelists voted in favor of this approach [S19]). The panelists agreed that the decision to administer olaparib followed by a CDK4/6 inhibitor should be made on an individual patient basis, after counseling the patient on the toxicity risks and unclear efficacy benefits. During discussions, emphasis was given to the importance of starting olaparib treatment within 12 weeks after prior adjuvant therapy, and that the optimal olaparib treatment duration is 1 year. These points are based on evidence from the Olympia trial, which showed that 12 months’ adjuvant treatment with olaparib improved OS compared with placebo in patients with BRCA1/2 high-risk HER2− early BC [25]. However, in this trial, patients receiving olaparib treatment required frequent dose interruptions to manage toxicities (dose interruption reported in 44.5%) [25]. Combining pembrolizumab with olaparib was deemed an acceptable option in the high-risk adjuvant setting, with 94% of panelists in agreement (S20).

Metastatic BC

Hormone Receptor-Positive, Human Epidermal Growth Factor Receptor 2-Low BC
In Saudi Arabia, several questions remain around the optimal management of patients with HR+ metastatic BC with low/ultra-low HER2 expression. In particular, if patients fail endocrine therapy, some clarity is required regarding starting targeted ADC therapy after the failure of endocrine therapy and the sequencing of ADC treatment.
Consensus was clear on T-DXd being the preferred ADC after failure of endocrine therapy and one line of chemotherapy in this patient population (S1). There is good evidence from the DESTINYBreast-04 trial to support this position, which showed significant improvements in progression-free survival (PFS) and OS with T-DXd versus physician’s choice of chemotherapy [26]. Low HER2 expression was defined in DESTINYBreast-04 as a score of 1+ on immunohistochemical (IHC) analysis or an IHC score of 2+ and negative results on in situ hybridization [26]. The positive impact of T-DXd on PFS in patients with metastatic HR+ BC with ultra-low HER2 expression was demonstrated in the DESTINYBreast-06 trial, which defined ultra-low expression as a score of 0 on IHC membrane staining [27]. Consensus was not reached on whether T-DXd should be started immediately after failure of two prior lines of endocrine therapy in HER2-low/-ultra-low disease (S2). Some panelists had concerns regarding the indication to start T-DXd immediately, since patients started T-DXd without intervening chemotherapy in the DESTINYBreast-06 trial [27], but in DESTINYBreast-04, chemotherapy was offered prior to initiating T-DXd [26]. The current lack of mature survival data for patients in DESTINYBreast-06 further complicates the decision-making process between these two treatment strategies [27].
Regarding the sequencing of ADCs in the HR+, HER2-low metastatic BC population, many panelists (65%) expressed a preference for first-line T-DXd followed by sacituzumab govitecan (S4), and were in agreement against giving first-line sacituzumab govitecan followed by T-DXd (S3). In theory, some benefit could be obtained from administering T-DXd before/after sacituzumab govitecan because these two ADCs have different targets (HER2 and trophoblast cell surface antigen-2 [TROP-2], respectively) but the same chemotherapy payload.
Consensus could not be reached on the ‘sandwich’ approach of ADC therapy followed by conventional chemotherapy and then ADC therapy again (S5). The lack of data on ADCs regarding mechanisms of resistance and overcoming resistance through exploiting a bystander effect limited the panelists’ ability to make firm recommendations in this regard. Although the findings of small observational studies presented at academic congresses show that one or more systemic therapy agents can be used between two ADCs and that PFS after the second ADC is approximately  2.5–3.7 months [28–30]; this small clinical benefit should be balanced against the financial toxicity of a ‘sandwich’ approach.
The panelists’ discussion acknowledged the lack of data for sequential ADCs and ADC ‘sandwich’ protocols, and that the PFS of any sequential ADC therapy is likely to yield PFS results similar to current standard protocols.

Human Epidermal Growth Factor Receptor 2-Low TNBC
Clear consensus was achieved in clarifying that sacituzumab govitecan is the preferred ADC for patients with TNBC after the failure of one line of chemotherapy (S6). This consensus statement also included patients with HER2-low expression; therefore, T-DXd is another potential treatment option. Evidence for the use of sacituzumab govitecan and for T-DXd is available from the ASCENTl [31] and DESTINYBreast-04 [26] trials, respectively. Toxicity from sacituzumab govitecan after one or two lines of prior chemotherapy can potentially be lessened with the addition of filgrastim (a granulocyte colony-stimulating factor) and loperamide to the ADC treatment regimen, as suggested by the results of the PRIMED trial [32]. Importantly, patients entering ASCENT had received two to four prior lines of chemotherapy, whereas the consensus statement S6 specifies that sacituzumab govitecan may be started after only one line of prior chemotherapy (i.e., can be used in the second-line setting). This consensus statement is consistent with the approved indication for sacituzumab govitecan in the USA, which states that patients must have received at least two prior lines of chemotherapy, of which only one of these must have been for metastatic disease [33]. In conclusion, sacituzumab govitecan can be used earlier (i.e., as second-line treatment) in patients who have received neoadjuvant/adjuvant therapy within 12 months of relapse.

Human Epidermal Growth Factor Receptor 2+ BC
Consensus statements relating to the management of patients with HER2+ metastatic BC in Saudi Arabia sought to clarify: (1) ADC sequencing in this patient population; (2) subsequent treatment steps after patients have received the CLEOPATRA regimen and have either brain metastases or leptomeningeal disease (LMD; S7 and S8); (3) how best to treat patients with progressive brain metastases but stable systemic disease (S9 and S10); and (4) treatment discontinuation in patients with a long-term complete metabolic response (S11).
The placebo-controlled phase 3 CLEOPATRA trial investigated the efficacy and safety of pertuzumab + trastuzumab and docetaxel in patients with HER2+ metastatic BC who had not received prior chemotherapy or biologic therapy [34–36]. The regimen was pertuzumab 420 mg (loading dose of 840 mg), trastuzumab 6 mg/kg (loading dose 8 mg/kg), and docetaxel 75–100 mg/m2, all given intravenously every 3 weeks until disease progression (pertuzumab or placebo and trastuzumab) or for a minimum of 6 cycles (docetaxel) [35]. Dose reductions of pertuzumab and trastuzumab were not allowed; docetaxel dose reductions of up to 25% were permitted.
The panelists agreed that T-DXd is the preferred second-line treatment following pertuzumab + trastuzumab (CLEOPATRA regimen) in patients with active brain metastases (S7). This consensus used the definition of active brain metastases used in DESTINYBreast-12: as either untreated brain metastases comprised of lesions ≤ 2 cm in size or brain metastases that had progressed after local central nervous system (CNS) therapy but with no clinical indication for immediate retreatment [37]. Based on the above, patients with active brain metastases should undergo comprehensive evaluation to determine the need for local treatment modalities, such as radiation or neurosurgical intervention, prior to initiation of systemic therapy. If no local therapy is indicated immediately, T-DXd should be considered the preferred option.
Similarly, T-DXd is a potential treatment option when patients with HER2+ metastatic BC have developed LMD after completion of the CLEOPATRA regimen (S8). In the phase 2 DEBBRAH trial, T-DXd was shown to be potentially effective for untreated LMD in patients previously treated for HER2+ (or HER2-low) metastatic BC with at least one prior line of chemotherapy and HER2-targeted and/or endocrine therapy as appropriate, in the advanced disease setting [38]. Median OS was 13.3 months (95% CI 5.7, not reached), suggesting the potential usefulness of targeted systemic therapy for patients with untreated LMD. Further, T-DXd demonstrated prolonged intra- and extracranial disease control in patients with HER2+ BC who had active brain metastases in the phase 2 TUXEDO-1 trial [39], and in patients with stable and active HER2 + brain metastases in the phase 3b/4 DESTINYBreast-12 trial [37]. Pooled data from the DESTINYBreast-01, 02 and 03 trials provide similar supportive data [40]. Data from ROSET-BM, a retrospective chart review, support the use of T-DXd for the treatment of patients with HER2+ metastatic BC with LMD and/or brain metastases (n = 104; n = 19 with LMD) [41–43]. Overall response rates in this latter study were high (55.7%) [43].
Alternate options for patients with HER2+ metastatic BC and LMD were discussed, including intrathecal trastuzumab therapy and tucatinib. Results from a phase 2 non-randomized study (TBCRC049; NCT03501979) in 17 patients with HER2+ BC suggest that tucatinib + trastuzumab + capecitabine is potentially effective for the first-line treatment of LMD [44–46], although enrollment was stopped early when tucatinib received US Food and Drug Administration (FDA) approval in April 2020 [45].
Consensus was not achieved for the remaining statements regarding patients with HER2+ metastatic BC (S9, S10, S11, S13, S15). There was no agreement on whether systemic therapy should be continued or changed in patients with stable systemic disease but progressive brain metastases (S9, S10), a common clinical scenario likely to be encountered by general/medical oncologists. If the patient is receiving trastuzumab, some panelists would immediately switch the systemic therapy to a brain-active treatment, such as tucatinib or T-DXd, even though these are associated with significant serious adverse effects, but others would avoid changing the systemic therapy from the relatively safe trastuzumab to preserve patient quality of life if the radiation oncologist and/or neurosurgeon expressed confidence in treating the brain metastases. Some panelists focused on the importance of establishing whether the progressive brain metastases are comprised of single or multiple lesions before deciding on a course of action. If there are multiple brain metastases, whole brain RT and the associated sequalae could be avoided by changing systemic therapy to a drug with high CNS penetration, whereas a single lesion could be treated with localized RT or stereotactic radio surgery. The ESMO metastatic BC guidelines suggest that when local intervention for progressive brain metastases is indicated, clinicians should consider continuing systemic therapy unchanged unless the progressive disease is outside the CNS [6]. If local therapy for progressive brain metastases is not indicated, then patients should receive T-DXd or trastuzumab plus tucatinib and capecitabine [6].
Despite a change in the consensus statement after the first round of voting, consensus could not be achieved on when to initiate discussion with the patient on discontinuing treatment once they have achieved a stable response to long-term maintenance therapy, especially given the lack of existing consensus guidelines (S11). It was confirmed that a ‘complete response’ (CR) implied a complete radiologic response. However, there was some concern expressed about differentiating between patients with CR who have a single bone metastasis (for example) versus oligometastatic disease versus widely spread metastatic disease. There was also discussion around why 3 years was selected as the time point in this statement, versus 4 or 5 years. Data from prospective, well-designed clinical trials on the optimal trastuzumab treatment duration in responders (i.e., when to discontinue) are currently lacking. This was made evident by the discordant views of the panelists on this question. One panelist reporting having observed disease recurrence in patients who discontinue long-term trastuzumab treatment. Another explained this has been referred to as ‘HER2 addiction’ that can occur with long-term HER2 inhibitor treatment, and that as soon treatment is stopped, disease relapse can occur. For example, in a small retrospective study from Finland, although the majority of patients (n = 17/21; 81%) discontinued HER2 inhibitor therapy after a long-lasting tumor response, and survival in metastatic disease was superior in patients who had CR or near-CR before discontinuation [47], almost one-half (47.6%) subsequently experienced disease progression after stopping HER2 inhibitor therapy, and treatment had to be reinitiated [47]. In contrast, a small Turkish retrospective study of patients with metastatic HER2+ BC found that only 13.3% of patients (n = 2/15) who discontinued trastuzumab-based treatment experienced disease recurrence during follow-up (median [range] 32 [11–66] months) [48]. A similar low rate of disease recurrence was reported in an Italian retrospective study [49]. These contradictory results from various observational studies underscore the urgent need for further research on the optimal duration of HER2 inhibitor treatment in patients who achieve CR. Results from the ongoing phase 2 STOP-HER2 trial conducted in patients with HER2+ metastatic BC may assist in providing evidence to support clinical decision-making in this regard [50].

Other Clinical Scenarios
The panelists agreed that sequencing of approved ADCs is an acceptable practice in the metastatic setting, confirming during the meeting that this statement is applicable to patients with HER2-low BC (S12). In contrast, no consensus was reached on whether chemotherapy should be offered before offering another ADC after progression on a previous ADC treatment, with only just over half of all panelists voting ‘Yes’ in support of this consensus statement (S13).
In patients with metastatic BC of any subtype, the overwhelming majority of panelists (95%) agreed that local therapy is an acceptable option for patients who develop oligoprogression, while maintaining the current systemic therapy regimen (S14). One of the first prospective phase 2 trials to address this question was recently published by clinicians from the Princess Margaret Cancer Centre in Toronto, Canada (the RADIANT trial), which enrolled patients with breast (n = 23) or genitourinary or gastrointestinal cancer (n = 47) [51]. Metastatic disease was identified at only one site in most patients (80%), and at two to four sites in the remaining 20%. The median (interquartile range [IQR]) duration of follow-up for the entire study population was 12.3 (8.2–21.6) months. Forty-one percent of BC patients had a change of systemic therapy at 1 year (cumulative incidence 40.5%; 95% CI 23.9–68.5)—due to progressive disease in all cases—and of the patients who required a change in systemic therapy, the median time from stereotactic body RT (SBRT) to this change was 6 (IQR 4.6–11.6) months [51]. Median PFS in the BC cohort was 6.5 months (95% CI 4.1–24.0). Safety data were reassuring, since this aspect is of most concern when giving SBRT for oligometastatic disease while continuing systemic therapy (particularly targeted therapy) [52]. No patient in the RADIANT trial changed systemic therapy due to systemic treatment-related toxicity, there were no grade 4 or 5 adverse events, and the cumulative incidence of acute grade ≥ 2 toxicity in the BC cohort was 8.7% (95% CI 2.2–33.7) [51]. Less encouraging results for SBRT as local therapy + systemic therapy were observed for the BC cohort (n = 47) in another prospective phase 2 trial (the CURB trial) in patients with oligoprogressive metastatic cancer (overall population n = 106) [53]. However, in CURB, most of the BC cohort had ≥ 2 sites of oligoprogression (88% of patients) and one-third had TNBC [53], while in RADIANT, most patients had only one site of oligoprogression, and only 9% of the BC cohort had TNBC (87% had HR+, HER− disease) [51]. Further research is required to identify selection criteria for patients in whom local therapy such as SBRT + systemic treatment may be appropriate to manage oligoprogressive disease.
Interestingly, Delphi consensus statements from the European Society for Radiotherapy (ESTRO) and the European Organization for Research and Treatment of Cancer (EORTC) Oligocare consortium recommend interrupting several (but not all) types of systemic immune or targeted therapy during metastasis-directed SBRT [54], whereas ESMO guidelines for metastatic BC broadly recommend combining systemic therapy with RT and/or surgery for oligometastatic disease, while individualizing treatment [6]. However, it should be noted that there are uncertainties and potential risks associated with concomitant radiation therapy and newer systemic therapies, such as targeted therapies, as outlined by ESTRO [55]. The panel recommends cautious use, close multidisciplinary collaboration, and possible washout intervals between systemic agents and radiation, although specific washout durations depend on the drugs used and the clinical context.
No consensus was reached on the final statement (S15) on lengthening of the dosing interval of maintenance denosumab (a receptor activator of nuclear factor-κ [RANK] ligand inhibitor) in patients with metastatic BC and bone metastases who are ineligible for bisphosphonate treatment (e.g., those with renal disease). Typically, subcutaneous denosumab is administered monthly in patient with bone metastases, consistent with current ESMO and American Society of Clinical Oncology (ASCO) guidelines [6, 56]. Although this is a convenient dosing regimen, it is associated with a considerable cost burden in Saudi Arabia, so clinicians may seek to reduce this financial toxicity by lengthening the dosing interval to once every 3 months (at 12-week intervals). One of the panelists stated that in their clinical practice, patients are switched to 3-monthly dosing after 1 year of treatment with the standard monthly denosumab regimen. However, there is limited evidence supporting lengthening the denosumab dosing interval. A retrospective study in patients with bone metastases (n = 116/432 [33.6%] with BC) found no difference in time to first skeletal-related event between patients with an extended dosing schedule (5–11-week or ≥ 12-week intervals) versus a short interval (< 5-week interval) [57]. Results from a randomized prospective study in 263 patients with bone metastases also suggested that a 12-week dosing interval of bone targeted agents, including denosumab, is a reasonable treatment option [58]. This prospective study enrolled patients with BC (60.8%) or castration-resistant prostate cancer (39.2%), approximately half of whom were bone-agent naïve at study entry (51.7%). Overall during the study, more patients received denosumab (56.3%) than zoledronate (24.0%) or pamidronate (19.8%). In the BC cohort, 1-year symptomatic skeletal events (SSE)-free survival was not statistically significantly different between 4-weekly and 12-weekly dosing (82.2% in the 4-weekly group vs 68.0% in the 12-weekly group; p = 0.051) [58]. Evidence for the bisphosphonate zoledronic acid is available from a large prospective randomized trial in patients with metastatic cancers (n = 1822), including BC (n = 855), showing that the skeletal event incidence over 2 years is not significantly increased after changing the dosing interval from 4 to 12 weeks [59].
The panelists also briefly mentioned concerns relating to osteonecrosis of the jaw (ONJ), an uncommon but serious adverse event that occurs with both denosumab and bisphosphonate treatment. Of note, the incidence of ONJ is reported to be higher with denosumab than zoledronic acid (9.6% vs 4.8%; p = 0.009) [60].

Limitations
While it is hoped that the consensus statements developed and presented in this paper will have significant impact on the treatment of BC in Saudi Arabia, it is recognized that they are based on the considerations and clinical expertise and experience of a select group of oncologists. Further, the statements may not be generalizable to clinical settings in other countries, particularly those with healthcare systems that differ from that of Saudi Arabia.

Conclusions
This modified Delphi process of an expert panel of oncologists from across Saudi Arabia has developed consensus statements on the management of early or metastatic BC. These statements highlight the nuances required in the interpretation of the latest clinical data in clinical decision-making, as well as the importance of real-world experience. By focusing on aspects of care for which there is less or contradictory evidence, particularly for clinical scenarios that oncologists are likely to encounter, it is hoped that these consensus statements will help standardize care, support decision-making, and ensure that Saudi oncologists have clear, locally relevant recommendations for managing complex BC cases, where high-level evidence is limited.

Supplementary Information
Below is the link to the electronic supplementary material.