The Cancer Drugs Fund ( CDF ) Panel has assessed the application for Eribulin ( Halaven ) to enter the CDF as an option for patients with locally advanced or metastatic HER-2 negative breast cancer who have already received at least one line of chemotherapy for advanced disease and been previously treated with an anthracycline and a taxane.
This application was in treatment pathway terms for the use of Eribulin as a 2nd, 3rd or 4th line therapy of advanced breast cancer.
The CDF Panel was aware that this specific indication for Eribulin had not been assessed by NICE. The manufacturer informed the CDF Panel that a NICE appraisal was expected in 2016 to cover this indication.
The Panel observed that NICE TA 250 did not recommend Eribulin for the treatment of locally advanced or metastatic breast cancer which had progressed after at least 2 chemotherapy regimens for advanced disease.
The manufacturer submitted as its main evidence the published paper in the Journal of Clinical Oncology ( JCO ) of a randomised trial ( study 301 ) which randomised 1102 patients with locally advanced or metastatic breast cancer to receive Eribulin vs Capecitabine.
Patients could have had up to 3 prior chemotherapy regimens and up to 2 previous chemotherapy regimens for advanced / metastatic disease but had to have received both an anthracycline and a taxane.
Patients could have received all, some or none of these regimens as adjuvant therapy, the Panel noting that 58% of patients had received some form of chemotherapy as adjuvant treatment.
Patients previously treated with Capecitabine were excluded from the study.
The trial was stratified according to HER-2 status and geographic regions of the world.
Coprimary endpoints of the study were overall survival ( OS ) and independently-assessed progression free survival ( PFS ).
The CDF Panel noted that 20% of patients received the study therapy as 1st line treatment of advanced / metastatic disease, 52% as 2nd line treatment and 27% as 3rd line therapy.
The Panel observed that 97.5% of patients were of WHO performance status 0 or 1.
In respect of HER-2 receptor status, 68.5% were HER-2 negative, 15% were HER-2 positive and 16% were of unknown HER-2 status.
Quality of life was regularly assessed using the EORTC QLQ C30 and breast module BR23, the principal pre-specified outcome being global QOL measured at week 6.
None of the primary endpoints were met in study 301. Median independently-assessed progression-free survival was 4.1 months for Eribulin versus 4.2 months for Capecitabine ( hazard ratio [ HR ] 1.08, 95% confidence interval [ CI ] 0.93-1.25, p=0.30 ).
Median investigator-assessed progression-free survival was 4.2 months for Eribulin versus 4.1 mo for Capecitabine ( HR=0.98, 95% CI 0.86-1.11, p=0.74 ).
Overall survival was 15.9 vs 14.5 months, respectively ( HR=0.88, 95% CI 0.77-1.00, p=0.056 ) and was not significantly different in the intention to treat analysis, the latter determined by pre-defined criteria for statistical significance.
The CDF Panel noted that cross over was allowed for Eribulin patients to receive Capecitabine ( and this occurred in 50% ) but not for Capecitabine patients to be treated with Eribulin until the very end of the study ( and this occurred in 0.4% ).
The Panel thus considered that this disparity in cross over potentially favoured survival in the Eribulin arm.
The CDF Panel was aware that the extension to the marketing authorisation granted approval for the use of Eribulin in patients previously treated with an anthracycline and a taxane who had received one or more lines of chemotherapy for advanced disease. It noted that this extension did not apply to 20% of the patients entered into the 301 study.
It observed that the European Public Assessment Report ( EPAR ) stated that there was not a statistically significant difference in overall survival in those patients treated with 1 or more regimens for advanced / metastatic disease ( HR=0.87, 95% CI 0.76-1.01 ), nor for progression-free survival ( HR=0.96, 95% CI 0.83-1.11 ).
The manufacturer’s application was for Eribulin to be funded by the CDF only for patients with HER-2 negative breast cancer who had received one or more lines of chemotherapy for their advanced / metastatic disease.
The median overall survival figures for all the HER-2 negative patients in study 301 ( 755 patients ) were 15.9 months for Eribulin vs 13.5 months for Capecitabine ( difference [ delta ] 2.4 months, HR=0.84, 95% CI 0.72-0.98, p=0.03 ).
For progression-free survival in the HER-2 negative group, the figures were stated by the manufacturer to be 4.0 vs 4.0 months ( HR=1.04, 95% CI 0.87-1.23, p=ns ) although the CDF Panel noted that the European Public Assessment Report indicated that the progression-free survival in the HER-2 negative group was 4.1 vs 3.9 months ( HR=0.95, 95% CI 0.81-1.10, p=ns ), respectively.
For the further subgroup of HER-2 negative patients who had only received 1 line of treatment for advanced / metastatic disease ( ie the subgroup of patients in the trial which mirrored the manufacturer’s application to the CDF ), the median overall survival figures were 15.9 vs 13.4 months ( delta 2.5 months, HR=0.81, 95% CI 0.68-0.97, p=0.048 ).
In terms of quality of life ( QOL ), more than 95% of data was available at baseline for both arms and completion rates decreased similarly in both arms as patients continued through the study.
QOL scores improved over time in both arms of the 301 study, there being no statistically significant difference between the two groups.
Adverse events were assessed according to the National Cancer Institute Common Terminology Criteria for Adverse Events. All grade treatment-emergent adverse events reported as treatment-related were seen in 85% of patients in the Eribulin arm versus 77% in the Capecitabine arm.
Treatment-emergent adverse effects of grade greater than or equal to 3 occurred in 65% vs 46% respectively. Serious adverse effects were observed in 17.5% of patients in the Eribulin arm versus 21.1% in the Capecitabine group.
Adverse effects which led to discontinuation of treatment were seen in 7.9% vs 10%, to reduction in dose in 32% vs 32% and to treatment delay in 32% vs 36%, respectively.
Life-threatening adverse effects occurred in 2.2% vs 3.5%, fatal adverse effects in 4.8% vs 6.6% and grade 3 or 4 febrile neutropenia in 2 vs 0.9%, respectively.
G-CSF ( granulocyte-colony stimulating factor ) use was noted in 15% vs 4%, respectively.
In terms of specific side-effects, some were clearly greater for Eribulin ( all grade alopecia 35% vs 4%, peripheral neuropathy 28% vs 14% ) whereas in other toxicities, the reverse was the case ( all grade diarrhoea 14% vs 29%, hand-foot syndrome 0% vs 45% ).
The manufacturer also submitted evidence from another randomised trial ( EMBRACE trial, study 305 ) as supportive evidence. This randomised 762 patients with locally advanced or metastatic breast cancer previously treated with at least 2 lines of chemotherapy to receive Eribulin vs active treatment of physician’s choice ( a variety of treatments, nearly all chemotherapies ). 68% of patients had previously received treatment with Capecitabine.
Median progression free survival as assessed by the investigators was significantly greater with Eribulin ( 3.6 vs 2.2 months, delta 1.4 months, HR=0.76, 95% CI 0.64-0.90, p=0.002 ) whereas independently assessed progression-free survival was not statistically different ( 3.7 vs 2.2 months, HR=0.87, 95% CI 0.71-1.05, p=0.137 ), respectively.
Updated overall survival was significantly superior with Eribulin ( 13.2 vs 10.5 months, delta 2.7 months, HR=0.81, 95% CI 0.67-0.96, p=0.014 ), respectively.
Safety was assessed according to the National Cancer Institute Common Terminology Criteria for Adverse Events. The following all grade toxicities were increased in the Eribulin arm: fatigue ( 54% vs 40% ), alopecia ( 45% vs 10% ), peripheral neuropathy ( 35% vs 16% ), arthralgia / myalgia ( 22% vs 12% ), weight loss ( 21% vs 14% ) and grade 3 and 4 neutropenia ( 45% vs 21% ), respectively. Febrile neutropenia occurred in 5% vs 2%, respectively. However, palmar-plantar erythrodysaesthesia was less common in Eribulin patients ( 1% vs 14% ).
Quality of life was not collected in this study.
The median duration of treatment in the Eribulin arm was 3.9 months. No mention is made in the EMBRACE trial publication as to whether cross over was allowed or not.
The CDF Panel examined the overall results results for HER-2 negative and positive patient sugbroups in studies 301 and 305. In study 301 and in the HER-2 negative group, overall survival was 15.9 months for Eribulin versus 13.5 months for Capecitabine ( HR=0.84, 95% CI 0.72-0.98, p=0.03 ) and for the HER-2 positive population, the corresponding figures were 14.3 vs 17.1 months ( HR=0.97, 0.69-1.36, p=0.84 ).
In study 305 ( EMBRACE ) and in the HER-2 negative group, overall survival was 13.4 months for Eribulin versus 10.5 months for treatment of physician’s choice ( HR=0.85, 0.70-1.04, p=0.11 ) and for the HER-2 positive population, the corresponding figures were 11.8 vs 8.9 months ( HR=0.59, 0.39-0.91, p=0.015 ).
The CDF Panel was concerned that the HRs for the HER-2 positive groups were very different in these two studies and could have been achieved by chance and thus gave doubt as to the relevance of the analysis by HER-2 status.
The manufacturer stated that a pooled analysis of study 301 and the EMBRACE study had been requested by the European Medicines Agency to determine whether the observed benefit of Eribulin in HER-2 negative disease was consistent.
In the application to the CDF, the pooled analysis was stated as demonstrating a difference in overall survival of 3.1 months between Eribulin and non-Eribulin chemotherapy ( 15.1 vs 12.0 months respectively, HR=0.84, 95% CI 0.74-0.96, p=0.011 ) and of 0.8 months in progression-free survival ( 3.7 vs 2.9 months, HR=0.84, 95% CI 0.73-0.96, p=0.009 ).
The CDF Panel noted that the publication of these pooled analyses for the HER-2 negative population in the paper in Breast Cancer Research and Treatment showed different figures to the ones supplied by the manufacturer. In this paper, the pooled analysis was stated as demonstrating a difference in overall survival of 2.9 months between Eribulin and non-Eribulin chemotherapy ( 15.2 vs 12.3 months respectively, HR=0.82, 95% CI 0.72-0.93, p=0.002 ) and of 0.7 months in progression-free survival ( 3.7 vs 3.0 months, HR=0.84, 95% CI 0.74-0.95, p=0.006 ).
The CDF Panel noted the disparities in the figures quoted by the manufacturer and those in the published paper although observed that the actual differences between these two sets of figures were small.
The authors of the paper describing the pooled analysis stated that women with HER-2 negative disease are among those who may obtain benefit from Eribulin and the pooled analysis offered useful insights into specific subgroups of patients with advanced breast cancer who may gain a greater benefit in terms of overall survival than others from Eribulin treatment.
The CDF Panel observed that this publication of the pooled analysis also stated that in the results of the HER-2 positive group ( which had much smaller numbers ), the HR for overall survival was also 0.82 ( 95% CI 0.62-1.06, p=0.135 ) yet the HR for progression-free survival was 1.02 ( 95% CI 0.78-1.34, p=0.87 ).
In addition, the CDF noted that this unplanned pooled analysis had not only pooled patients from different places in the treatment pathway and thus with different degrees of pre-treatment with chemotherapy ( eg 70% of patients in study 301 had received 1 or less lines of chemotherapy for advanced disease whereas the figure for the EMBRACE study was 1% ) but also pooled patients treated with different comparators ( Capecitabine in study 301, treatment of physician’s choice in the EMBRACE study ).
The CDF Panel recognised that one of the key issues in its consideration of this Eribulin application to the CDF was the robustness of the HER-2 negative subgroup in terms of the panel’s decision making.
The CDF Panel observed that the 301 study had been stratified for HER-2 status. It noted that the 301 study publication stated that pre-specified exploratory analyses were conducted to assess an effect of Eribulin according to HER-2 status. The 301 study publication stated that although a possible benefit according to HER-2 status was suggested for overall survival, an interaction test showed no benefit for Eribulin when comparing patients with HER-2 negative disease and all other patients ( HER-2 positive and unknown HER-2 status ).
The CDF panel then examined the European Public Assessment Report ( EPAR ) for Eribulin which had also considered the issue of subgroups in the 301 study according to receptor status ( HER-2 receptor and oestrogen receptor [ ER ] ).
The European Medicine Agency’s relevant considerations on receptor status are set out below: In subgroup analysis by receptor status, a larger relative overall survival treatment effect of Eribulin compared with Capecitabine was seen in the ER negative, hormone receptor negative and HER2 negative pre-specified subgroups, compared to their complementary receptor positive subgroups, and most of all in the combined triple-negative subgroup of patients: overall survival HR 0.70 ( 95% CI: 0.545; 0.91 ) vs. 0.93 ( n.s ) in the complementary set.
The progression-free survival subgroup analyses showed similar trends of a relatively better outcome for Eribulin vs Capecitabine in the triple negative subgroup and in the HER2-negative subgroup.
However, opposite trends were observed in the updated overall survival subgroup analysis of the registration Study 305, with better HR point estimates for ER positive, PgR [ progesterone receptor ] positive and HER2 positive subgroups, compared with their negative counterparts, and better overall survival HR estimates for the non-triple negative subgroup compared with the triple-negative; 0.77 ( 95% CI: 0.63; 0.95 ) vs 0.89 ( 95% CI 0.60; 1.32 ).
For progression-free survival, the updated HRs for Study 305, provided in responses to questions, were very similar in triple negative vs the complementary set, 0.79 vs 0.78.
The conflicting findings between Study 301 and Study 305 with regard to receptor subgroups weaken the observations made in study 301.
Furthermore, pooled data ( although questionable from a methodological point of view ) for studies 305 and 301 indicated similar effects for HER2-positive and -negative patients, with pooled overall survival HR 0.82 ( 95% CI: 0.62; 1.06 ) and 0.84 ( 95% CI: 0.74; 0.95 ), respectively; and pooled progression-free survival HR 1.02 ( 95% CI: 0.78, 1.34 ) and 0.84 ( 95% CI: 0.74, 0.95 ), respectively.
Based on the data from both phase 3 trials in breast cancer, Studies 305 and 301, it is concluded that no relevant difference in efficacy has been shown between the HER2 positive and negative patient subgroups, or between triple negative vs non- triple negative subgroups.
The authors of the pooled analysis noted that the fact that women with triple-negative breast cancer may derive a particularly strong benefit from Eribulin is worthy of further investigation. In vitro and in vivo data suggest that Eribulin treatment may promote transition of triple-negative breast cancer cells from a mesenchymal to an epithelial phenotype, and that this is coupled with a decrease in the ability of these cells to migrate and therefore metastasize.
The CDF Panel noted that this data referred to triple negative breast cancer and the pooled analysis ( with all the Panel’s reservations about its robustness ) showed that there was no longer a statistically significant overall survival benefit when the subgroup with HER-2 negative but oestrogen receptor positive was examined ( HR=0.86, 95% CI 0.74-1.01, p=0.06 ).
The CDF Panel was informed that the manufacturer had formulated Eribulin into a new 3 ml vial which results in less potential wastage and thus a more efficient administration of drug.
The CDF Panel recognised that the marketing authorisation for Eribulin was for Eribulin to be an option for patients with locally advanced or metastatic breast cancer who have already received at least one line of chemotherapy for advanced disease and been previously treated with an anthracycline and taxane.
It noted that the HRs for progression-free survival and overall survival were similar according to the intention to treat analyses in study 301 and for the remaining patients once those patients had been excluded who had received trial therapy as their 1st line treatment of advanced / metastatic disease.
The CDF Panel concluded that it was not satisfied that separating the HER-2 negative patients from the HER-2 positive patients was a robust procedure and noted that the European Medicines Agency ( EMA ) had also come to the same conclusion.
It also did not regard the pooled analyses of studies 301 and 305 as appropriate given the heterogeneity of the patient populations in the two trials and the different comparators used.
The CDF Panel thus regarded the intention to treat analyses as the best data on which to score the clinical benefit of Eribulin even though the licensed indication had restricted approval to the patients who had progressed after at least one chemotherapy regimen for advanced disease.
The CDF Panel again noted that the EMA when making this restriction from the clinical trial population into its licensed indication had made no similar action in regard to HER-2 status.
Finally, the CDF Panel considered whether Eribulin fulfilled the CDF criteria for unmet need. The CDF panel was aware of treatments given for advanced breast cancer and noted that active comparators were used in both study 301 ( Capecitabine ) and study 305 ( treatment of physicians’ choice ).
Eribulin was therefore one of a number of therapies that could be used in the management of advanced / metastatic breast cancer and thus did not fulfil the definition of unmet need as set out in the CDF Standard Operating Procedures.
The Panel did not regard Eribulin as being a step change in the management of breast cancer given its failure to demonstrate an intention to treat superiority when compared with Capecitabine.
The CDF Panel thus scored the clinical impact of Eribulin in patients who could have had up to 3 prior chemotherapy regimens and at least one previous chemotherapy regimen for advanced / metastatic disease but had to have received both an anthracycline and a taxane as being according to the results of study 301: for progression-free survival, the score was 0; for overall survival, the score was 0; for quality of life, the score was 1; for toxicity, the score was 0; for unmet need, the score was 0; and thus the total was 1B.
In conclusion, the CDF Panel considered the clinical benefits of Eribulin in patients with HER-2 negative locally advanced or metastatic breast cancer who have progressed after at least one line of chemotherapy to be insufficiently robust to justify entry into the CDF.
The CDF Panel also considered the clinical benefits of Eribulin in patients with locally advanced or metastatic breast cancer who have progressed after at least one line of chemotherapy to be insufficient to justify entry into the CDF. ( Xagena )
Source: Cancer Drugs Fund ( UK ), 2015