HALAVEN is indicated for the treatment of patients with metastatic breast cancer who have previously received at least two chemotherapeutic regimens for the treatment of metastatic disease.  Prior therapy should have included an anthracycline and a taxane in either the adjuvant or metastatic setting [see Clinical Studies (14.1) ].

HALAVEN is indicated for the treatment of patients with unresectable or metastatic liposarcoma who have received a prior anthracycline-containing regimen [see Clinical Studies (14.2) ].

The recommended dose of HALAVEN is 1.4 mg/m² administered intravenously over 2 to 5 minutes on Days 1 and 8 of a 21-day cycle.

The recommended dose of HALAVEN in patients with mild hepatic impairment (Child-Pugh A) is 1.1 mg/m² administered intravenously over 2 to 5 minutes on Days 1 and 8 of a 21-day cycle [ see Use in Specific Populations (8.6) ].

The recommended dose of HALAVEN in patients with moderate hepatic impairment (Child-Pugh B) is 0.7 mg/m² administered intravenously over 2 to 5 minutes on Days 1 and 8 of a 21-day cycle [ see Use in Specific Populations (8.6) ].

The recommended dose of HALAVEN in patients with moderate or severe renal impairment (creatinine clearance (CLcr) 15-49 mL/min) is 1.1 mg/m² administered intravenously over 2 to 5 minutes on Days 1 and 8 of a 21-day cycle [ see Use in Specific Populations (8.7) ].

Assess for peripheral neuropathy and obtain complete blood cell counts prior to each dose.

Recommended dose delays

• Do not administer HALAVEN on Day 1 or Day 8 for any of the following:
  - ANC < 1,000/mm³
  - Platelets < 75,000/mm³
  - Grade 3 or 4 non-hematological toxicities.
  
• The Day 8 dose may be delayed for a maximum of 1 week.
  - If toxicities do not resolve or improve to ≤ Grade 2 severity by Day 15, omit the dose.
  - If toxicities resolve or improve to ≤ Grade 2 severity by Day 15, administer HALAVEN at a reduced dose and initiate the next cycle no sooner than 2 weeks later.

Recommended dose reductions

• If a dose has been delayed for toxicity and toxicities have recovered to Grade 2 severity or less, resume HALAVEN at a reduced dose as set out in Table 1.
  
• Do not re-escalate HALAVEN dose after it has been reduced.

Table 1: Recommended Dose Reductions

Event Description	Recommended   HALAVEN Dose
Permanently reduce the 1.4 mg/m 2 HALAVEN dos e for any of the following:	1.1 mg/m2
ANC <500/mm3 for >7 days
ANC <1,000 /mm3 with fever or infection
Platelets <25,000/mm3
Platelets <50,000/mm3 requiring transfusion
Non-hematologic Grade 3 or 4 toxicities
Omission or delay of Day 8 HALAVEN dose in previous cycle for toxicity
Occurrence of any event requiring permanent dose reduction while receiving 1.1 mg/m2	0.7 mg/m2
Occurrence of any event requiring permanent dose reduction while receiving 0.7 mg/m2	Discontinue HALAVEN
ANC = absolute neutrophil count. Toxicities graded in accordance with National Cancer Institute (NCI) Common Terminology Criteria for Adverse Events (CTCAE) version 3.0.

Aseptically withdraw the required amount of HALAVEN from the single-dose vial and administer undiluted or diluted in 100 mL of 0.9% Sodium Chloride Injection, USP.

Do not dilute in or administer through an intravenous line containing solutions with dextrose. Do not administer in the same intravenous line concurrent with the other medicinal products.

Store undiluted HALAVEN in the syringe for up to 4 hours at room temperature or for up to 24 hours under refrigeration at 4°C (40°F). Store diluted solutions of HALAVEN for up to 4 hours at room temperature or up to 24 hours under refrigeration at 4°C (40°F).

Discard unused portions of the vial.

In Study 1, severe neutropenia (ANC < 500/mm3) lasting more than one week occurred in 12% (62/503) of patients with metastatic breast cancer, leading to discontinuation in <1% of patients.  Febrile neutropenia (fever ≥38.5°C with Grade 3 or 4 neutropenia) occurred in 5% (23/503) of patients; two patients (0.4%) died from complications of febrile neutropenia [see Adverse Reactions (6.1) ].

In Study 1, patients with alanine aminotransferase (ALT) or aspartate aminotransferase (AST) > 3 × ULN (upper limit of normal) experienced a higher incidence of Grade 4 neutropenia and febrile neutropenia than patients with normal aminotransferase levels. Patients with bilirubin > 1.5 × ULN also had a higher incidence of Grade 4 neutropenia and febrile neutropenia.

In Study 2, severe neutropenia (ANC < 500/mm3) lasting more than one week occurred in 12% (26/222) of patients with liposarcoma or leiomyosarcoma. Febrile neutropenia occurred in 0.9% of patients treated with HALAVEN and fatal neutropenic sepsis in 0.9% [see Adverse Reactions (6.1)].

Monitor complete blood counts prior to each dose; increase the frequency of monitoring in patients who develop Grade 3 or 4 cytopenias. Delay administration of HALAVEN and reduce subsequent doses in patients who experience febrile neutropenia or Grade 4 neutropenia lasting longer than 7 days [ see Dosage and Administration (2.2) ]. Clinical studies of HALAVEN did not include patients with baseline neutrophil counts below 1,500/mm³.

In Study 1, Grade 3 peripheral neuropathy occurred in 8% (40/503) of patients, and Grade 4 in 0.4% (2/503) of patients with metastatic breast cancer (MBC). Peripheral neuropathy was the most common toxicity leading to discontinuation of HALAVEN (5% of patients; 24/503) in Study 1.  Neuropathy lasting more than one year occurred in 5% (26/503) of patients. Twenty-two percent (109/503) of patients developed a new or worsening neuropathy that had not recovered within a median follow-up duration of 269 days (range 25-662 days).

In Study 2, Grade 3 peripheral neuropathy occurred in 3.1% (7/223) of HALAVEN-treated patients. Peripheral neuropathy led to discontinuation of HALAVEN in 0.9% of patients. The median time to first occurrence of peripheral neuropathy of any severity was 5 months (range: 3.5 months to 9 months).  Neuropathy lasting more than 60 days occurred in 58% (38/65) of patients.  Sixty three percent (41/65) had not recovered within a median follow-up duration of 6.4 months (range: 27 days to 29 months).

Monitor patients closely for signs of peripheral motor and sensory neuropathy. Withhold HALAVEN in patients who experience Grade 3 or 4 peripheral neuropathy, until resolution to Grade 2 or less [ see   Dosage and Administration (2.2) ].

Based on findings from an animal reproduction study and its mechanism of action, HALAVEN can cause fetal harm when administered to a pregnant woman.  There are no adequate and well-controlled studies of HALAVEN in pregnant women. In animal reproduction studies, eribulin mesylate caused embryo-fetal toxicity when administered to pregnant rats during organogenesis at doses below the recommended human dose. Advise pregnant women of the potential risk to a fetus. Advise females of reproductive potential to use effective contraception during treatment with HALAVEN and for at least 2 weeks following the final dose.  Advise males with female partners of reproductive potential to use effective contraception during treatment with HALAVEN and for 3.5 months following the final dose [see Use in Specific Populations (8.1) ].

In an uncontrolled open-label ECG study in 26 patients, QT prolongation was observed on Day 8, independent of eribulin concentration, with no QT prolongation observed on Day 1. ECG monitoring is recommended if therapy is initiated in patients with congestive heart failure, bradyarrhythmias, drugs known to prolong the QT interval, including Class Ia and III antiarrhythmics, and electrolyte abnormalities. Correct hypokalemia or hypomagnesemia prior to initiating HALAVEN and monitor these electrolytes periodically during therapy. Avoid HALAVEN in patients with congenital long QT syndrome.

Because clinical trials are conducted under widely varying conditions, the adverse reaction rates observed in the clinical trials of a drug cannot be directly compared to rates in other clinical trials and may not reflect the rates observed in clinical practice.

The following adverse reactions are discussed in detail in other sections of the labeling:

• Neutropenia [ see Warnings and Precautions (5.1) ]
  
• Peripheral neuropathy [see Warnings and Precautions (5.2)]
  
• QT prolongation [see Warnings and Precautions (5.4) ]

In clinical trials, HALAVEN has been administered to 1963 patients including 467 patients exposed to HALAVEN for 6 months or longer. The majority of the 1963 patients were women (92%) with a median age of 55 years (range: 17 to 85 years). The racial and ethnic distribution was White (72%), Black (4%), Asian (9%), and other (3%).

Metastatic Breast Cancer

The most common adverse reactions (≥25%) reported in patients receiving HALAVEN were neutropenia, anemia, asthenia/fatigue, alopecia, peripheral neuropathy, nausea, and constipation. The most common serious adverse reactions reported in patients receiving HALAVEN were febrile neutropenia (4%) and neutropenia (2%). The most common adverse reaction resulting in discontinuation of HALAVEN was peripheral neuropathy (5%).

The adverse reactions described in Table 2 were identified in 750 patients treated in Study 1 [ see Clinical Studies (14.1) ]. In Study 1, patients were randomized (2:1) to receive either HALAVEN (1.4 mg/m²  on Days 1 and 8 of a 21-day cycle) or single agent treatment chosen by their physician (control group). A total of 503 patients received HALAVEN and 247 patients in the control group received therapy consisting of chemotherapy [total 97% (anthracyclines 10%, capecitabine 18%, gemcitabine 19%, taxanes 15%, vinorelbine 25%, other chemotherapies 10%)] or hormonal therapy (3%). The median duration of exposure was 118 days for patients receiving HALAVEN and 63 days for patients receiving control therapy. Table 2 reports the most common adverse reactions occurring in at least 10% of patients in either group.

Table 2: Adverse Reactions^a with a Per-Patient Incidence of at Least 10% in Study 1

Adverse Reactions	HALAVEN n=503		Control Group n=247
	All Grades	≥ Grade 3	All Grades	≥ Grade 3
Blood and l ymphatic s ystem d isorders b
Neutropenia	82%	57%	53%	23%
Anemia	58%	2%	55%	4%
Nervous system disorders
Peripheral neuropathyc	35%	8%	16%	2%
Headache	19%	<1%	12%	<1%
General disorders
Asthenia/Fatigue	54%	10%	40%	11%
Pyrexia	21%	<1%	13%	<1%
Mucosal inflammation	9%	1%	10%	2%
Gastrointestinal disorders
Nausea	35%	1%	28%	3%
Constipation	25%	1%	21%	1%
Vomiting	18%	1%	18%	1%
Diarrhea	18%	0	18%	0
Musculoskeletal and connective tissue disorders
Arthralgia/Myalgia	22%	<1%	12%	1%
Back pain	16%	1%	7%	2%
Bone pain	12%	2%	9%	2%
Pain in extremity	11%	1%	10%	1%
Metabolism and nutrition disorders
Decreased weight	21%	1%	14%	<1%
Anorexia	20%	1%	13%	1%
Respiratory, thoracic, and mediastinal disorders
Dyspnea	16%	4%	13%	4%
Cough	14%	0	9%	0
Skin and subcutaneous tissue disorders
Alopecia	45%	NAd	10%	NAd
Infections
Urinary Tract Infection	10%	1%	5%	0
a  adverse reactions were graded per National Cancer Institute Criteria for Adverse Events version 4.0.  b  based upon laboratory data. c  includes peripheral neuropathy, peripheral sensorimotor neuropathy, peripheral motor neuropathy, polyneuropathy, peripheral sensory neuropathy, and paraesthesia. d  not applicable; (grading system does not specify > Grade 2 for alopecia).

Cytopenias: Grade 3 neutropenia occurred in 28% (143/503) of patients who received HALAVEN in Study 1, and 29% (144/503) of patients experienced Grade 4 neutropenia. Febrile neutropenia occurred in 5% (23/503) of patients; two patients (0.4%) died from complications of febrile neutropenia. Dose reduction due to neutropenia was required in 12% (62/503) of patients and discontinuation was required in <1% of patients.  The mean time to nadir was 13 days and the mean time to recovery from severe neutropenia (<500/mm³) was 8 days.  Grade 3 or greater thrombocytopenia occurred in 1% (7/503) of patients.  G-CSF (granulocyte colony-stimulating factor) or GM-CSF (granulocyte–macrophage colony-stimulating factor) was used in 19% of patients who received HALAVEN.  

Peripheral Neuropathy:  In Study 1, 17% of enrolled patients had Grade 1 peripheral neuropathy and 3% of patients had Grade 2 peripheral neuropathy at baseline. Dose reduction due to peripheral neuropathy was required by 3% (14/503) of patients who received HALAVEN. Four percent (20/503) of patients experienced peripheral motor neuropathy of any grade and 2% (8/503) of patients developed Grade 3 peripheral motor neuropathy.

Liver Function Test Abnormalities:  Among patients with Grade 0 or 1 ALT levels at baseline, 18% of HALAVEN-treated patients experienced Grade 2 or greater ALT elevation. One HALAVEN-treated patient without documented liver metastases had concomitant Grade 2 elevations in bilirubin and ALT; these abnormalities resolved and did not recur with re-exposure to HALAVEN.

Less Common Adverse Reactions:   The following additional adverse reactions were reported in ≥5% to <10% of the HALAVEN-treated group:

• Eye Disorders:  increased lacrimation
  
• Gastrointestinal Disorders:  dyspepsia, abdominal pain, stomatitis, dry mouth
  
• General Disorders and Administration Site Conditions:   peripheral edema
  
• Infections and Infestations:  upper respiratory tract infection
  
• Metabolism and Nutrition Disorders:  hypokalemia
  
• Musculoskeletal and Connective Tissue Disorders:    muscle spasms, muscular weakness
  
• Nervous System Disorders:  dysgeusia, dizziness
  
• Psychiatric Disorders:  insomnia, depression
  
• Skin and Subcutaneous Tissue Disorders:  rash

Liposarcoma

The safety of HALAVEN was evaluated in Study 2, an open-label, randomized, multicenter, active-controlled trial, in which patients were randomized (1:1) to receive either HALAVEN 1.4 mg/m² on Days 1 and 8 of a 21-day cycle or dacarbazine at doses of 850 mg/m² (20%), 1000 mg/m² (64%), or 1200 mg/m² (16%) every 3 weeks.  A total of  223 patients received HALAVEN and 221 patients received dacarbazine. Patients were required to have received at least two prior systemic chemotherapy regimens.  The trial excluded patients with pre-existing ≥ Grade 3 peripheral neuropathy, known central nervous system metastasis, elevated serum bilirubin or significant chronic liver disease, history of myocardial infarction within 6 months, history of New York Heart Association Class II or IV heart failure, or cardiac arrhythmia requiring treatment. The median age of the safety population in Study 2 was 56 years (range: 24 to 83 years); 67% female; 73% White, 3% Black or African American, 8% Asian/Pacific Islander, and 15% unknown; 99% received prior anthracycline-containing regimen; and 99% received ≥ 2 prior regimens. The median duration of exposure was 2.3 months (range: 21 days to 26 months) for patients receiving HALAVEN [ see Clinical Studies (14.2) ].

The most common adverse reactions (≥25%) reported in patients receiving HALAVEN were fatigue, nausea, alopecia, constipation, peripheral neuropathy, abdominal pain, and pyrexia.  The most common (≥5%) Grade 3-4 laboratory abnormalities reported in patients receiving HALAVEN were neutropenia, hypokalemia, and hypocalcemia. The most common serious adverse reactions reported in patients receiving HALAVEN were neutropenia (4.9%) and pyrexia (4.5%). Permanent discontinuation of HALAVEN for adverse reactions occurred in 8% of patients. The most common adverse reactions resulting in discontinuation of HALAVEN were fatigue and thrombocytopenia (0.9% each).  Twenty-six percent of patients required at least one dose reduction. The most frequent adverse reactions that led to dose reduction were neutropenia (18%) and peripheral neuropathy (4.0%).

Table 3 summarizes the incidence of adverse reactions occurring in at least 10% of patients in the HALAVEN-treated arm in Study 2.

Table 3: Adverse Reactions^a Occurring in ≥10% (all Grades) of Patients Treated on the HALAVEN arm and at a Higher Incidence than in the Dacarbazine Arm (Between Arm Difference of ≥5% for All Grades or ≥2% for Grades 3 and 4) (Study 2)^b

Adverse Reaction	HALAVEN n =22 3		Dacarbazine n =22 1
	All Grades	Grades 3-4	All Grades	Grades 3-4
Nervous s ystem d isorders
Peripheral Neuropathyc	29%	3.1%	8%	0.5%
Headache	18%	0%	10%	0%
General d isorders
Pyrexia	28%	0.9%	14%	0.5%
G astrointestinal   d isorders
Constipation	32%	0.9%	26%	0.5%
Abdominal paind	29%	1.8%	23%	4.1%
Stomatitis	14%	0.9%	5%	0.5%
Skin and s ubcutaneous t issue d isorders
Alopecia	35%	NAe	2.7%	NAe
Infections
Urinary tract infection	11%	2.2%	5%	0.5%
a Adverse reactions were graded per National Cancer Institute Criteria for Adverse Events version 4.03 (NCI CTCAE v4.03). b Safety data from one study site enrolling six patients were excluded. c  Includes peripheral neuropathy, peripheral sensorimotor neuropathy, peripheral motor neuropathy, polyneuropathy, peripheral sensory neuropathy, and paraesthesia. d Includes abdominal pain, upper abdominal pain, lower abdominal pain, abdominal discomfort. e Not applicable; (grading system does not specify > Grade 2 for alopecia).

Other clinically important adverse reactions occurring in ≥10% of the HALAVEN-treated patients were:

• Gastrointestinal Disorders: nausea (41%); vomiting (19%), diarrhea (17%)
  
• General Disorders:  asthenia/fatigue (62%); peripheral edema (12%)
  
• Metabolism and Nutrition Disorders: decreased appetite (19%)
  
• Musculoskeletal and Connective Tissue Disorders:    arthralgia/myalgia (16%); back pain (16%)
  
• Respiratory Disorders:  cough (18%)

Less Common Adverse Reactions: The following additional clinically important adverse reactions were reported in ≥5% to <10% of the HALAVEN-treated group:

• Blood and Lymphatic System Disorders: thrombocytopenia
  
• Eye Disorders: increased lacrimation
  
• Gastrointestinal Disorders: dyspepsia
  
• Metabolism and Nutrition Disorders: hyperglycemia
  
• Musculoskeletal and Connective Tissue Disorders: muscle spasms, musculoskeletal pain
  
• Nervous System Disorders: dizziness, dysgeusia
  
• Psychiatric Disorders: insomnia, anxiety
  
• Respiratory, Thoracic, and Mediastinal Disorders: oropharyngeal pain
  
• Vascular Disorders: hypotension

Table 4: Laboratory Abnormalities Occurring in ≥10% (all Grades) of Patients Treated on the HALAVEN arm and at a Higher Incidence than in the Dacarbazine Arm (Between Arm Difference of ≥5% for All Grades or ≥2% for Grades 3 and 4)^a (Study 2)^†

Laboratory Abnormality	H alaven		Dacarbazine
	All Grades	Grades 3 - 4	All Grades	Grades 3 – 4
Hematology
Anemia	70%	4.1%	52%	6%
Neutropenia	63%	32%	30%	8.9%
Chemistry
Increased alanine aminotransferase (ALT)	43%	2.3%	28%	2.3%
Increased aspartate aminotransferase (AST)	36%	0.9%	16%	0.5%
Hypokalemia	30%	5.4%	14%	2.8%
Hypocalcemia	28%	5%	18%	1.4%
Hypophosphatemia	20%	3.2%	11%	1.4%
a  Each test incidence is based on the number of patients who had both baseline and at least one on-study measurement and at least 1 grade increase from baseline. Halaven group (range 221-222) and dacarbazine group (range 214-215). †   Laboratory results were graded per NCI CTCAE v4.03.

The following adverse drug reactions have been identified during post-approval of HALAVEN.  Because these reactions are reported voluntarily from a population of uncertain size, it is not always possible to reliably estimate their frequency or establish a causal relationship to drug exposure.

• Blood and Lymphatic System Disorders:  lymphopenia
  
• Gastrointestinal Disorders:  pancreatitis
  
• Hepatobiliary Disorders:  hepatotoxicity
  
• Immune System Disorders:  drug hypersensitivity
  
• Infections and Infestations:  pneumonia, sepsis/neutropenic sepsis
  
• Metabolism and Nutrition Disorders:  hypomagnesemia, dehydration
  
• Respiratory, thoracic and mediastinal disorders:  interstitial lung disease
  
• Skin and Subcutaneous Tissue Disorders:  pruritus, Stevens-Johnson syndrome, toxic epidermal necrolysis

No drug-drug interactions are expected with CYP3A4 inhibitors, CYP3A4 inducers or P-glycoprotein (P-gp) inhibitors. Clinically meaningful differences in exposure (AUC) were not observed in patients with advanced solid tumors when HALAVEN was administered with or without ketoconazole (a strong inhibitor of CYP3A4 and a P-gp inhibitor) and when HALAVEN was administered with or without rifampin (a CYP3A4 inducer) [see Clinical Pharmacology (12.3) ].   

Eribulin does not inhibit CYP1A2, CYP2C9, CYP2C19, CYP2D6, CYP2E1 or CYP3A4 enzymes or induce CYP1A2, CYP2C9, CYP2C19 or CYP3A4 enzymes at relevant clinical concentrations. Eribulin is not expected to alter the plasma concentrations of drugs that are substrates of these enzymes [see Clinical Pharmacology (12.3) ].

Risk Summary

Based on findings from an animal reproduction study and its mechanism of action, HALAVEN can cause fetal harm when administered to a pregnant woman [see Clinical Pharmacology (12.1) ]. There are no available data on the use of HALAVEN during pregnancy. In an animal reproduction study, eribulin mesylate caused embryo-fetal toxicity when administered to pregnant rats during organogenesis at doses below the recommended human dose [see Data].  Advise pregnant women of the potential risk to a fetus.

The estimated background risks of major birth defects and miscarriage for the indicated populations are unknown. In the U.S. general population, the estimated background risk of major birth defects and miscarriage in clinically-recognized pregnancies is 2% to 4% and 15% to 20%, respectively.

Data

Animal Data

In an embryo-fetal developmental toxicity study, pregnant rats received intravenous infusion of eribulin mesylate during organogenesis (Gestation Days 8, 10, and 12) at doses approximately 0.04, 0.13, 0.43 and 0.64 times the recommended human dose, based on body surface area. Increased abortion and severe fetal external or soft tissue malformations, including the absence of a lower jaw and tongue, or stomach and spleen, were observed at doses 0.64 times the recommended human dose of 1.4 mg/m² based on body surface area. Increased embryo-fetal death/resorption, reduced fetal weights, and minor skeletal anomalies consistent with developmental delay were also reported at doses at or above a maternally toxic dose of approximately 0.43 times the recommended human dose.

Risk Summary

There is no information regarding the presence of eribulin mesylate or its metabolites in human milk, the effects on the breastfed infant, or the effects on milk production. No lactation studies in animals were conducted.  Because of the potential for serious adverse reactions in breastfed infants from eribulin mesylate, advise women not to breastfeed during treatment with HALAVEN and for 2 weeks after the final dose.

Contraception

Females

Based on findings from an animal reproduction study and its mechanism of action, HALAVEN can cause fetal harm when administered to a pregnant woman [see Use in Specific Populations (8.1) ].  Advise females of reproductive potential to use effective contraception during treatment with HALAVEN and for at least 2 weeks following the final dose.

Males

Based on its mechanism of action, advise males with female partners of reproductive potential to use effective contraception during treatment with HALAVEN and for 3.5 months following the final dose.

Infertility  

Males 

Based on animal data, HALAVEN may result in damage to male reproductive tissues leading to impaired fertility of unknown duration [see Nonclinical Toxicology (13.1) ].

The safety and effectiveness of HALAVEN in pediatric patients have not been established.

The safety and effectiveness of HALAVEN alone or in combination with irinotecan in pediatric patients were assessed but not established in three open-label studies (NCT02171260, NCT03441360, and NCT03245450) in 77 pediatric patients aged 2 to <17 years with relapsed or refractory solid tumors and lymphomas, excluding central nervous system tumors. No new safety signals were observed in these studies. The pharmacokinetics (PK) of eribulin were within range of values of adult patients with metastatic liposarcoma or other tumors given the same dose per body surface area.

Study 1 did not include sufficient numbers of subjects with metastatic breast cancer aged 65 years and older to determine whether they respond differently from younger subjects. Of the 827 subjects who received the recommended dose and schedule of HALAVEN in clinical studies with advanced breast cancer, 15% (121/827) were 65 and older, and 2% (17/827) patients were 75 and older. No overall differences in safety were observed between these subjects and younger subjects.

Clinical studies of HALAVEN did not include a sufficient number of subjects in Study 2 aged 65 years and older to determine whether they respond differently from younger subjects.

Administration of HALAVEN at a dose of 1.1 mg/m² to patients with mild hepatic impairment and 0.7 mg/m² to patients with moderate hepatic impairment resulted in similar exposure to eribulin as a dose of 1.4 mg/m² to patients with normal hepatic function. Therefore, a lower starting dose of 1.1 mg/m² is recommended for patients with mild hepatic impairment (Child-Pugh A) and of 0.7 mg/m² is recommended for patients with moderate hepatic impairment (Child-Pugh B). HALAVEN was not studied in patients with severe hepatic impairment (Child-Pugh C) [see Dosage and Administration (2.1) , Clinical Pharmacology (12.3) ] .  

For patients with moderate or severe renal impairment (CLcr 15-49 mL/min), reduce the starting dose to 1.1 mg/m² [see Dosage and Administration (2.1) , Clinical Pharmacology (12.3) ] .

Eribulin inhibits the growth phase of microtubules without affecting the shortening phase and sequesters tubulin into nonproductive aggregates. Eribulin exerts its effects via a tubulin-based antimitotic mechanism leading to G₂/M cell-cycle block, disruption of mitotic spindles, and, ultimately, apoptotic cell death after prolonged mitotic blockage.

In addition, eribulin treatment of human breast cancer cells caused changes in morphology and gene expression as well as decreased migration and invasiveness in vitro. In mouse xenograft models of human breast cancer, eribulin treatment was associated with increased vascular perfusion and permeability in the tumor cores, resulting in reduced tumor hypoxia, and changes in the expression of genes in tumor specimens associated with a change in phenotype.

Cardiac Electrophysiology

The effect of HALAVEN on the QTc interval was assessed in an open-label, uncontrolled, multicenter, single-arm dedicated QT trial. A total of 26 patients with solid tumors received 1.4 mg/m² of HALAVEN on Days 1 and 8 of a 21-day cycle. A delayed QTc prolongation was observed on Day 8, with no prolongation observed on Day 1.  The maximum mean QTcF change from baseline (95% upper confidence interval) was 11.4 (19.5) ms.

The pharmacokinetics (PK) of eribulin is linear with a mean elimination half-life of approximately 40 hours, a mean volume of distribution of 43 L/m² to 114 L/m² and mean clearance of 1.16 L/hr/m² to 2.42 L/hr/m² over the dose range of 0.25 mg/m² to 4.0 mg/m².  The human plasma protein binding of eribulin at concentrations of 100 ng/mL to 1,000 ng/mL ranges from 49% to 65%. Eribulin exposure after multiple dosing is comparable to that following a single dose. No accumulation of eribulin is observed with weekly administration.

Elimination

Metabolism

Unchanged eribulin was the major circulating species in plasma following administration of ¹⁴C-eribulin to patients. Metabolite concentrations represented <0.6% of parent compound, confirming that there are no major human metabolites of eribulin.  Cytochrome P450 3A4 (CYP3A4) negligibly metabolizes eribulin in vitro.

Excretion

Eribulin is eliminated primarily in feces unchanged. After administration of ¹⁴C-eribulin to patients, approximately 82% of the dose was eliminated in feces and 9% in urine.  Unchanged eribulin accounted for approximately 88% and 91% of total eribulin in feces and urine, respectively. 

Specific Populations

Age, Sex , and Race /Ethnicity:  Based on a population pharmacokinetic analysis, no clinically meaningful differences in the pharmacokinetics of eribulin were observed based on age, sex, or race.

Hepatic Impairment

In a study evaluating the effect of hepatic impairment on the PK of eribulin, eribulin exposures increased by 1.8-fold in patients with mild hepatic impairment (Child-Pugh A; n=7) and by 2.5-fold in patients with moderate (Child-Pugh B; n=5) hepatic impairment as compared to patients with normal hepatic function (n=6).  Administration of HALAVEN at a dose of 1.1 mg/m² to patients with mild hepatic impairment and 0.7 mg/m² to patients with moderate hepatic impairment resulted in similar exposure to eribulin at a dose of 1.4 mg/m² to patients with normal hepatic function [see Dosage and Administration (2.1) , Use in Specific Populations (8.6) ] .

Renal Impairment

In a study evaluating the effect of renal impairment on the PK of eribulin, patients with moderate (CLcr 30-49 mL/min; n=7) and severe renal impairment (CLcr 15-29 mL/min; n=6) had 1.5-fold higher eribulin dose-normalized exposures compared to that in patients with normal renal function (CLcr ≥ 80 mL/min; n=6).  There were no clinically meaningful changes in patients with mild renal impairment (CLcr 50-79 mL/min; n=27) [see Dosage and Administration (2.1) , Use in Specific Populations (8.7) ] .

Drug Interaction Studies

Effect of Strong Inhibitors or Inducers of CYP3A4 on Eribulin:  The effect of a strong CYP3A4 inhibitor and a P-gp inhibitor, ketoconazole, on the PK of eribulin was studied in a crossover trial of 12 patients with advanced solid tumors. No clinically relevant PK interaction was observed when HALAVEN was administered with or without ketoconazole (the geometric mean ratio of the AUC: 0.97; 90% CI: 0.83, 1.12).

The effect of a CYP3A4 inducer, rifampin, on the PK of eribulin was studied in a crossover trial of 14 patients with advanced solid tumors. No clinically relevant PK interaction was observed when HALAVEN was administered with or without rifampin (the geometric mean ratio of the AUC: 1.10; 90 CI%: 0.91, 1.34).

Effect of Eribulin on CYP Substrates:  Eribulin shows no induction potential for CYP1A, CYP2B6, CYP2C9, CYP2C19, and CYP3A in primary human hepatocytes. Eribulin inhibits CYP3A4 activity in human liver microsomes, but it is unlikely that eribulin will substantially increase the plasma levels of CYP3A4 substrates. No significant inhibition of CYP1A2, CYP2C9, CYP2C19, CYP2D6, or CYP2E1 was detected with eribulin concentrations up to 5 μM in pooled human liver microsomes. In vitro drug interaction studies indicate that eribulin does not inhibit drugs that are substrates of these enzymes and it is unlikely that eribulin will affect plasma levels of drugs that are substrates of CYP enzymes.

Effect of Transporters on Eribulin:  In vitro data suggest that eribulin at clinically relevant concentrations is a substrate of P-gp, but is not a substrate of breast cancer resistance protein (BCRP), multidrug resistance proteins (MRP2, MRP4), bile salt extrusion pump (BSEP), organic anion transporting polypeptides (OATP1B1, OATP1B3), organic anion transporters (OAT1, OAT3), organic cation transporters (OCT1, OCT2), or multidrug and toxin extrusion 1 (MATE1).

Effect of Eribulin on Transporters:  In vitro data suggest that eribulin at clinically relevant concentrations may inhibit P-gp, but does not inhibit BCRP, OATP1B1, OCT1, OAT1, OAT3, or MATE1.

Carcinogenicity studies have not been conducted with eribulin mesylate.  Eribulin mesylate was not mutagenic in in vitro bacterial reverse mutation assays (Ames test). Eribulin mesylate was positive in mouse lymphoma mutagenesis assays, and was clastogenic in an in vivo rat bone marrow micronucleus assay.

Fertility studies have not been conducted with eribulin mesylate in humans or animals; however, nonclinical findings in repeat-dose dog and rat toxicology studies suggest that male fertility may be compromised by treatment with eribulin mesylate.  Rats exhibited testicular toxicity (hypocellularity of seminiferous epithelium with hypospermia/aspermia) following dosing with eribulin mesylate at or above 0.43 times the recommended human dose (based on body surface area) given once weekly for 3 weeks, or at or above 0.21 times the recommended human dose (based on body surface area) given once weekly for 3 out of 5 weeks, repeated for 6 cycles. Testicular toxicity was also observed in dogs given 0.64 times the recommended human dose (based on body surface area) weekly for 3 out of 5 weeks, repeated for 6 cycles.

Study 1 was an open-label, randomized, multicenter trial of 762 patients with metastatic breast cancer who received at least two chemotherapeutic regimens for the treatment of metastatic disease and experienced disease progression within 6 months of their last chemotherapeutic regimen. Patients were required to receive prior anthracycline- and taxane-based chemotherapy for adjuvant or metastatic disease. Patients were randomized (2:1) to receive HALAVEN (n=508) or a single agent therapy selected prior to randomization (control arm, n=254). Randomization was stratified by geographic region, HER2/neu status, and prior capecitabine exposure. HALAVEN was administered at a dose of 1.4 mg/m2 on Days 1 and 8 of a 21-day cycle. HALAVEN-treated patients received a median of 5 cycles (range: 1 to 23 cycles) of therapy.  Control arm therapy consisted of 97% chemotherapy (26% vinorelbine, 18% gemcitabine, 18% capecitabine, 16% taxane, 9% anthracycline, 10% other chemotherapy), and 3% hormonal therapy. The main efficacy outcome was overall survival.

Patient demographic and baseline characteristics were comparable between the treatment arms. The median age was 55 (range: 27 to 85 years) and 92% were White. Sixty-four percent of patients were enrolled in North America/Western Europe/Australia, 25% in Eastern Europe/Russia, and 11% in Latin America/South Africa.  Ninety-one percent of patients had a baseline ECOG performance status of 0 or 1. Tumor prognostic characteristics, including estrogen receptor status (positive: 67%, negative: 28%), progesterone receptor status (positive: 49%, negative: 39%), HER2/neu receptor status (positive: 16%, negative: 74%), triple negative status (ER-, PR-, HER2/neu-: 19%), presence of visceral disease (82%, including 60% liver and 38% lung) and bone disease (61%), and number of sites of metastases (greater than two: 50%), were also similar in the HALAVEN and control arms. Patients received a median of four prior chemotherapy regimens in both arms.

In Study 1, a statistically significant improvement in overall survival was observed in patients randomized to the HALAVEN arm compared to the control arm (see Table 5). An updated, unplanned survival analysis, conducted when 77% of events had been observed (see Figure 1), was consistent with the primary analysis. In patients randomized to HALAVEN, the objective response rate by the RECIST criteria was 11% (95% CI: 8.6%, 14.3%) and the median response duration was 4.2 months (95% CI: 3.8, 5.0 months).

Table 5: Comparison of Overall Survival in HALAVEN and Control Arm - Study 1

Overall Survival	HALAVEN (n=508)	Control Arm (n=254)
Primary survival analysis
Number of deaths	274	148
Median, months (95% CI)	13.1 (11.8, 14.3)	10.6 (9.3, 12.5)
Hazard Ratio (95% CI)a	0.81 (0.66, 0.99)
P valueb	0.041
Updated survival analysis
Number of deaths	386	203
Median, months (95% CI)	13.2 (12.1, 14.4)	10.6 (9.2, 12.0)
CI = confidence interval a Based on Cox proportional hazards model stratified by geographic region, HER2 status, and prior capecitabine therapy. b Based on a log-rank test stratified by geographic region, HER2 status, and prior capecitabine therapy.

The efficacy and safety of HALAVEN were evaluated in Study 2, an open-label, randomized (1:1), multicenter, active-controlled trial.  Eligible patients were required to have unresectable, locally advanced or metastatic liposarcoma or leiomyosarcoma, at least two prior systemic chemotherapies (one of which must have included an anthracycline), and disease progression within 6 months of the most recent chemotherapy regimen.  Patients were randomized to HALAVEN 1.4 mg/m² administered intravenously on Days 1 and 8 of a 21-day cycle or to dacarbazine at a dose of 850 mg/m², 1000 mg/m², or 1200 mg/m² administered intravenously every 21 days (dacarbazine dose was selected by the investigator prior to randomization).  Treatment continued until disease progression or unacceptable toxicity.  Randomization was stratified by histology (liposarcoma or leiomyosarcoma), number of prior therapies (2 vs. > 2), and geographic region (U.S. and Canada vs. Western Europe, Australia, and Israel vs. Eastern Europe, Latin America, and Asia).  The major efficacy outcome measure was overall survival (OS). Additional efficacy outcome measures were progression-free survival (PFS) and confirmed objective response rate (ORR) as assessed by the investigator according to Response Evaluation Criteria in Solid Tumors (RECIST v1.1).  Patients in the dacarbazine arm were not offered HALAVEN at the time of disease progression.

A total of 446 patients were randomized, 225 to the HALAVEN arm and 221 to the dacarbazine arm. The median age was 56 years (range: 24 to 83); 33% were male; 73% were White; 44% had ECOG performance status (PS) 0 and 53% had ECOG PS 1; 68% had leiomyosarcoma and 32% had liposarcoma; 39% were enrolled in U.S. and Canada (Region 1) and 46% were enrolled in Western Europe, Australia, and Israel (Region 2); and 47% received more than two prior systemic chemotherapies.  The most common (>40%) prior systemic chemotherapies were doxorubicin (90%), ifosfamide (62%), gemcitabine (59%), trabectedin (50%), and docetaxel (48%).

Of the 143 patients with liposarcoma, the median age was 55 years (range: 32 to 83); 62% were male, 72% were White; 41% had ECOG PS of 0 and 53% had ECOG PS of 1; 35% were enrolled in Region 1 and 51% were enrolled in Region 2; and 44% received more than two prior systemic chemotherapies.  The distribution of subtypes of liposarcoma, based on local histologic assessment, were 45% dedifferentiated, 37% myxoid/round cell, and 18% pleomorphic.

Study 2 demonstrated a statistically significant improvement in OS in patients randomized to HALAVEN compared with dacarbazine (see Table 6).  There was no significant difference in progression-free survival in the overall population. Treatment effects of HALAVEN were limited to patients with liposarcoma based on pre-planned, exploratory subgroup analyses of OS and PFS (see Tables 6 and 7 and Figure 2). There was no evidence of efficacy of HALAVEN in patients with advanced or metastatic leiomyosarcoma in Study 2 (see Table 7).

Table 6: Efficacy Results for the Liposarcoma Stratum and All Patients* in Study 2^a

	Liposarcoma Stratum			All Patients *
	Halaven (n=71)		Dacarbazine (n=72)	Halaven (n=22 5 )	Dacarbazine (n=22 1 )
Overall s urvival
Deaths, n (%)	52 (73)		63 (88)	173 (77)	179 (81)
Median, months    (95% CI)	15.6 (10.2, 18.6)		8.4 (5.2, 10.1)	13.5 (11.1, 16.5)	11.3 (9.5, 12.6)
Hazard ratio (HR)    (95% CI)	0.51 (0.35, 0.75)			0.75 (0.61, 0.94)
Stratified log-rank p value	N/A†			0.011
Progression-free survival
Events, n (%)	57 (80)		59 (82)	194 (86)	185 (84)
Disease progression	53		52	180	170
Death	4		7	14	15
Median, months    (95% CI)	2.9 (2.6, 4.8)		1.7 (1.4, 2.6)	2.6 (2.0, 2.8)	2.6 (1.7, 2.7)
HR    (95% CI)	0.52  (0.35, 0.78)			0.86  (0.69, 1.06)
Objective response rate
Objective response rate (%)    (95% CI)	1.4  (0, 7.6)	0  (0, 4.2)		4.0  (1.8, 7.5)	5.0 (2.5, 8.7)

^a Efficacy data from one study site enrolling six patients were excluded.
*All patients = liposarcoma and leiomyosarcoma.
^†  N/A = not applicable

Table 7: Efficacy Results for the Leiomyosarcoma Stratum in Study 2^a

	Leiomyosarcoma Stratum
	Halaven (n=15 4 )	Dacarbazine (n=1 49 )
Overall s urvival
Deaths, n (%)	121 (79)	116 (78)
Median, months       (95% CI)	12.8 (10.3, 14.8)	12.3 (11.0, 15.1)
HR (95% CI)	0.90 (0.69, 1.18)
Progression-free survival
Events, n (%)	137 (89)	126 (85)
Disease progression	127	118
Death	10	8
Median, months    (95% CI)	2.2 (1.5, 2.7)	2.6 (2.2, 2.9)
HR (95% CI)	1.05 (0.81, 1.35)
Objective response rate (%)    (95% CI)	5.2 (2.3, 10)	7.4 (3.7, 12.8)

                 ^a Efficacy data from one study site enrolling six patients were excluded.