Expert UK consensus on the definition of high risk of recurrence in HER2-negative early breast cancer: A modified Delphi panel

BACKGROUND

There is currently no standardised definition for patients at high risk of recurrence of human epidermal growth factor receptor 2 (HER2)-negative early breast cancer (eBC; stages 1-3) after surgery. This modified Delphi panel aimed to establish expert UK consensus on this definition, separately considering hormone receptor (HR)-positive and triple-negative (TN) patients.

METHODS

Over three consecutive rounds, results were collected from 29, 24 and 22 UK senior breast cancer oncologists and surgeons, respectively. The first round aimed to determine key risk factors in each patient subgroup; subsequent rounds aimed to establish appropriate risk thresholds. Consensus was pre-defined as ≥70% of respondents.

RESULTS

Expert consensus was achieved on need to assess age, tumour size, tumour grade, number of positive lymph nodes, inflammatory breast cancer and risk prediction tools in all HER2-negative patients. There was additional agreement on use of tumour profiling tests and biomarkers in HR-positive patients, and pathologic complete response (pCR) status in TN patients. Thresholds for high recurrence risk were subsequently agreed. In HR-positive patients, these included age <35 years, tumour size >5 cm (as independent risk factors); tumour grade 3 (independently and combined with other high-risk factors); number of positive nodes ≥4 (independently) and ≥1 (combined). For TN patients, the following thresholds reached consensus, both independently and in combination with other factors: tumour size >2 cm, tumour grade 3, number of positive nodes ≥1.

CONCLUSIONS

The results may be a valuable reference point to guide recurrence risk assessment and decision-making after surgery in the HER2-negative eBC population.

Abstract P3-05-39: Expert consensus on the definition of high risk of recurrence in HER2-negative early breast cancer: a modified Delphi panel

PURPOSE: There is currently no standardised definition for patients at high risk of recurrence of HER2-negative early breast cancer (eBC, stages 1–3) after surgery. Recognising that the assessment of high risk is often multifactorial, the aim of this modified Delphi panel was to establish expert UK consensus on this definition, separately considering HR-positive and triple-negative (TN) patients. METHODS: A total of 45 UK-based clinicians, including breast cancer oncologists and surgeons, were invited to participate. The number of respondents in each of three rounds was 29, 24 and 22 respectively. Statements were developed using the results from a targeted literature review and the guidance of a lead clinician, and comprised free-text, single-choice or numerical formats. The first round aimed to determine which factors are currently used in clinical practice to assess risk of recurrence in the populations of interest. In the subsequent rounds, the objective was to establish thresholds indicative of high risk in a 10-year timeframe for each of the factors retained in Round 1. Between each round, statements were refined, considering the distribution of responses and free-text notes provided by participants. Consensus for single-choice questions was set at a pre-defined threshold of ≥70% of respondents. RESULTS: Consensus was achieved on the need to assess age, tumour size, tumour grade, number of positive nodes, presence of inflammatory breast cancer and one or more risk prediction tools to define high risk of recurrence in all HER2-negative patients. In HR-positive patients, there was agreement on the use of one or more tumour profiling tests and biomarkers to define high risk of recurrence. However, there was no consensus on biomarker use in TN patients, and support for specific biomarkers (such as Ki-67) was conflicting for both sub-populations based on the analysis of free-text notes. Similarly, while there was consensus on the use of pCR status/residual disease to indicate high risk in TN patients, this factor failed to reach consensus for the HR-positive sub-population. Germline BRCA status and menopausal status were not considered to be key factors for risk of recurrence in either biological subtype. In the second and third rounds, thresholds indicative of high recurrence risk were agreed; it should be noted that the free-text responses provided by the participants frequently highlighted that many of the factors should be considered along a continuous scale when assessing the risk of individual patients. In HR-positive patients, these thresholds included: age &lt; 35 years, tumour size &gt;5 cm (each when considered independently from other risk factors); tumour grade 3 (independently or in combination with other factors); number of positive lymph nodes ≥4 when considered independently or ≥1 in combination with other factors. For patients with TN tumours, the following thresholds reached consensus, whether considered independently or in combination with other factors: tumour size &gt;2 cm, tumour grade 3, number of positive lymph nodes ≥1. In several cases, however, no consensus could be reached on the appropriate threshold indicating high risk of recurrence. In the HR-positive sub-population, these included thresholds for age and tumour size, when considered in combination with other factors. In the TN sub-population, this included age, whether independently or in combination with other factors. CONCLUSIONS: The expert consensus reached in this panel highlights that an integrated model is important in assessing recurrence risk in eBC and that definitions of high risk differ according to biological subtype. The results may serve as a valuable reference point for clinicians to use in assessing risk of disease recurrence and in making treatment decisions after surgery in the HER2-negative eBC population. FUNDING: AstraZeneca UK Ltd. Writing support: Costello Medical.

Citation Format: Ellen R. Copson, Jean E. Abraham, Jeremy P. Braybrooke, Stuart A. McIntosh, Caroline O. Michie, Carlo Palmieri, Rebecca Roylance, Saiqa Spensley. Expert consensus on the definition of high risk of recurrence in HER2-negative early breast cancer: a modified Delphi panel [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P3-05-39.

Randomized multicenter trial of 3 weekly cabazitaxel versus weekly paclitaxel chemotherapy in the first-line treatment of HER2 negative metastatic breast cancer (MBC)

1008

Background: Paclitaxel is commonly used as first line chemotherapy for HER2 negative MBC. However, with response rates of 21.5-53.7% and a significant risk of peripheral neuropathy there is a need for more effective and better tolerated chemotherapy (CCT). Methods: This open label randomised (1:1) phase 2 trial compared 6 cycles of cabazitaxel (25 mg/m2) every 3 weeks, with weekly paclitaxel (80mg/m2) over 18 weeks as first line CCT. HER2 negative and performance status ≤1 patients were eligible. Patients on cabazitaxel received GCSF prophylaxis. Primary endpoint was Progression Free Survival (PFS) with 127 events required to detect a hazard ratio (HR) of 0.65 with 85% power. Secondary endpoints included objective response rate (ORR; RECIST 1.1), time to response (TTR), overall survival (OS), safety and tolerability and quality of life (QoL). Results: 158 patients were recruited from 14 UK hospitals (79 in each arm). Median age (range) was 56(34-81) in the cabazitaxel arm and 61(34-79) in the paclitaxel arm. 61% of patients were performance status 0. Median time on treatment was 15 weeks for both arms, but more patients on paclitaxel had a treatment delay (61% vs 39%) or dose reduction (37% vs 24%). Comparing cabazitaxel to paclitaxel after 146 PFS events, median PFS was 6.7 vs 5.8 months (HR 0.84; 95%CI 0.60–1.18, P = 0.3). There was no difference in OS, median 19.3 vs 20.0 months (HR 0.94; 95%CI 0.63-1.40, P = 0.7), ORR (42% vs 37%) or TTR (HR 1.09; 95%CI 0.68–1.74, P = 0.7). Grade ≥3 adverse events occurred in 42% of patients on cabazitaxel and 48% on paclitaxel. Diarrhoea, febrile neutropenia and nausea were the most common grade ≥3 events in the cabazitaxel arm with rates of 11%, 11% and 10% respectively compared to 1%, 1% and 0% in the paclitaxel arm. In the paclitaxel arm the top grade ≥3 events were lung infection and peripheral neuropathy, 6% and 5% respectively compared to 2.5% and 0% in the cabazitaxel arm. Peripheral neuropathy of any grade was reported by 55% of patients treated with paclitaxel vs 17% on cabazitaxel. Alopecia occurred in 41% of patients on paclitaxel compared to 27% on cabazitaxel. Adverse events leading to discontinuation were more frequent with paclitaxel (22%) than cabazitaxel (14%). Over the course of treatment, mean EQ5D single index utility score (+0.05; 95%CI 0.004-0.09, P = 0.03) and visual analogue scale score (+7.7; 95%CI 3.1-12.3, P = 0.001) were higher in the cabazitaxel arm compared to paclitaxel suggestive of better QoL on Cabazitaxel. Conclusions: 3 weekly cabazitaxel as first line chemotherapy in HER2 negative MBC does not significantly improve PFS compared to weekly paclitaxel, though it has a lower risk of peripheral neuropathy with better patient reported overall health outcomes. Cabazitaxel is safe and well tolerated for MBC and requires fewer hospital visits, an important consideration in the COVID pandemic and beyond. Clinical trial information: NCT03048942 .

Osteonecrosis of the Jaw and Oral Health–Related Quality of Life After Adjuvant Zoledronic Acid: An Adjuvant Zoledronic Acid to Reduce Recurrence Trial Subprotocol (BIG01/04)

Purpose

In patients with early breast cancer, adjuvant zoledronic acid (zoledronate) may reduce recurrence and improve survival. However, zoledronate is associated with the occasional development of osteonecrosis of the jaw (ONJ). We report on the frequency of ONJ and investigate oral health–related quality of life (Oral-QoL) in a large randomized trial (Adjuvant Zoledronic Acid to Reduce Recurrence [AZURE]).

Patients and Methods

Three thousand three hundred sixty women with stage II or III breast cancer were randomly assigned to receive standard adjuvant systemic therapy alone or with zoledronate administered at a dose of 4 mg for 19 doses over 5 years. All potential occurrences of ONJ were reported as serious adverse events and centrally reviewed. Additionally, we invited 486 study participants to complete the Oral Health Impact Profile-14 (OHIP-14) to assess Oral-QoL around the time the patients completed 5 years on study. Multivariable linear regression was used to calculate mean scores and 95% CIs in addition to identifying independent prognostic factors.

Results

With a median follow-up time of 73.9 months (interquartile range, 60.7 to 84.2 months), 33 possible cases of ONJ were reported, all in the zoledronate-treated patients. Twenty-six cases were confirmed as being consistent with a diagnosis of ONJ, representing a cumulative incidence of 2.1% (95% CI, 0.9% to 3.3%) in the zoledronate arm. Three hundred sixty-two patients (74%) returned the OHIP-14 questionnaire. Neither the prevalence nor severity of impacts on Oral-QoL differed significantly between zoledronate patients and control patients.

Conclusion

Adjuvant zoledronate used in the intensive schedule studied in the AZURE trial is associated with a low incidence of ONJ but does not seem to adversely affect Oral-QoL.

Functional outcome of malignant spinal cord compression treated with radiotherapy alone: A prospective analysis

e20623

Background: Malignant spinal cord compression (MSCC) is a major oncological complication. The management of Impending Malignant Spinal Cord Compression (IMSCC) remains unclear. Radiotherapy (RT) is often the sole management of both entities. Our aim was to prospectively evaluate the functional outcome of MSCC and IMSCC. Methods: All pts with MSCC and IMSCC treated by RT in our institution were screened for 2 national trials (ICORG 05–03/07–11). Non-eligible pts were prospectively evaluated and followed with assessment of mobility (modified Tomita scale) and sphincter function (continence/ incontinence/catheter) at baseline and 5 weeks posttreatment. Results: From 08/07 to 11/08, 54 pts [23 IMSCC, 31 MSCC] were followed. 31 pts were male. The median age was 60.5 y (30–86). The primary tumours were haematological [13 pts], lung [10], prostate [8], renal cell [6] and breast [5]. 51 pts had diagnostic MRI. The compression level was cervical in 6 pts, thoracic in 27, lumbosacral in 16, and 5 had multiple levels. 2D RT was used with varying radiation schedules: 20Gy/5fractions (f) [32 pts], 30Gy/10f [12] and other [10]. At baseline, normal mobility was found in 29 pts, mildly impaired mobility in 15 and immobility in 10. 7 pts had sphincter dysfunction. At 5 weeks, 38 pts (16 IMSCC, 22 MSCC) were evaluable (37 for mobility score, 38 for sphincter function). 10 pts had died and 6 were lost to follow up. Overall mobility was stable, improved or worse in 28 pts, 3 pts and 6 pts respectively. The mobility score was worse in 4 pts with IMSCC and 2 pts with MSCC. Improvement of mobility score was seen in 2 pts with MSCC. Among the 7 pts with baseline sphincter dysfunction, 2 pts improved. New sphincter dysfunction after RT was seen in 2 pts with IMSCC. Conclusions: The functional outcome of MSCC treated by RT alone remains poor, with minimal mobility and sphincter improvement. Functional outcome of pts with IMSCC is worrying with a worsening of mobility and/or occurrence of sphincter dysfunction in a large number of patients.

No significant financial relationships to disclose.

Clinical outcome for chemoradiotherapy in carcinoma of the cervix

AIMS

Two recent meta-analyses have shown a survival advantage for the addition of concurrent chemotherapy to radiotherapy in the treatment of cervical cancer. However, there is insufficient information available on late toxicity and few data from UK practice. The aims of this study were to examine treatment outcomes (survival and toxicity) in patients with cervical cancer treated with chemoradiation and to compare these with outcomes in patients treated with radiation alone.

MATERIALS AND METHODS

Between July 2000 and December 2003, 75 patients with cervical cancer were treated with chemoradiation. Case notes were reviewed retrospectively. Acute and late toxicity were recorded, with late toxicity graded using the Franco-Italian glossary. The median age was 47 years. All patients were staged with examination under anaesthesia and magnetic resonance imaging scans. Forty-two patients were treated with concurrent chemoradiation alone and 33 patients were treated with a combination of neoadjuvant and concurrent chemoradiation. This was due to waiting list problems. The chemotherapy used was cisplatin 40 mg/m(2) weekly with radiotherapy, (the neoadjuvant dose was 60 mg/m(2) 3 weekly). External beam radiotherapy was given to the pelvis (40-45 Gy/20 fractions/4 weeks) followed by low dose rate brachytherapy (22.5-32.5 Gy to point A). Patients who were unable to have brachytherapy were given an external beam boost (15-20 Gy/8-10 fractions).

RESULTS

The 3-year overall survival rate was 70%, with an estimated 5-year overall survival rate of 60%. The 3-year disease-free survival was 63.6%, with an estimated 5-year disease-free survival rate of 55%. Compared with the cohort of 183 patients from the Christie Hospital in a 1993 audit, there was a trend towards improved overall survival from 49 to 60% (P=0.06), which may become significant with longer follow-up. There were seven patients (9.3%) with grade 3 toxicity and no cases of grade 4 toxicity. In comparison with patients treated in the 1993 audit, the late toxicity rate has increased from 3.4 to 9.3%, but this was not statistically significant (P=0.14).

CONCLUSION

There was a trend towards improved survival with concurrent chemoradiation in this cohort of patients that may become significant with longer follow-up.