Ruxolitinib

Patients should be closely monitored and the dose titrated based on safety and efficacy.

In healthy subjects given ruxolitinib (50 mg single dose) following the potent CYP3A4 inducer rifampicin (600 mg daily dose for 10 days), ruxolitinib AUC was 70% lower than after administration of ruxolitinib alone. The exposure of ruxolitinib active metabolites was unchanged. Overall, the ruxolitinib pharmacodynamic activity was similar, suggesting the CYP3A4 induction resulted in minimal effect on the pharmacodynamics. However, this could be related to the high ruxolitinib dose resulting in pharmacodynamic effects near E_max. It is possible that in the individual patient, an increase of the ruxolitinib dose is needed when initiating treatment with a strong enzyme inducer.

In healthy subjects co-administration of ruxolitinib (10 mg single dose) with a dual CYP2C9 and CYP3A4 inhibitor, fluconazole, resulted in ruxolitinib C_max and AUC that were higher by 47% and 232%, respectively, than with ruxolitinib alone.

50% dose reduction should be considered when using medicinal products which are dual inhibitors of CYP2C9 and CYP3A4 enzymes (e.g. fluconazole). Avoid the concomitant use of ruxolitinib with fluconazole doses greater than 200 mg daily.

In patients with any hepatic impairment the recommended starting dose based on platelet count should be reduced by approximately 50% to be administered twice daily. Subsequent doses should be adjusted based on careful monitoring of safety and efficacy. Patients diagnosed with hepatic impairment while receiving ruxolitinib should have complete blood counts, including a white blood cell count differential, monitored at least every one to two weeks for the first 6 weeks after initiation of therapy with ruxolitinib and as clinically indicated thereafter once their liver function and blood counts have been stabilised. Ruxolitinib dose can be titrated to reduce the risk of cytopenia.

Ruxolitinib may inhibit P-glycoprotein and breast cancer resistance protein (BCRP) in the intestine. This may result in increased sytemic exposure of substrates of these transporters, such as dabigatran etexilate, ciclosporin, rosuvastatin and potentially digoxin. Therapeutic drug monitoring (TDM) or clinical monitoring of the affected substance is advised.

It is possible that the potential inhibition of P-gp and BCRP in the intestine can be minimised if the time between administrations is kept apart as long as possible.

Strong CYP3A4 inhibitors (such as, but not limited to, boceprevir, clarithromycin, indinavir, itraconazole, ketoconazole, lopinavir/ritonavir, ritonavir, mibefradil, nefazodone, nelfinavir, posaconazole, saquinavir, telaprevir, telithromycin, voriconazole):

In healthy subjects co-administration of ruxolitinib (10 mg single dose) with a strong CYP3A4 inhibitor, ketoconazole, resulted in ruxolitinib C_max and AUC that were higher by 33% and 91%, respectively, than with ruxolitinib alone. The half-life was prolonged from 3.7 to 6.0 hours with concurrent ketoconazole administration.

When administering ruxolitinib with strong CYP3A4 inhibitors the unit dose of ruxolitinib should be reduced by approximately 50%, to be administered twice daily. Patients should be closely monitored (e.g. twice weekly) for cytopenias and dose titrated based on safety and efficacy.

In healthy subjects co-administration of ruxolitinib (10 mg single dose) with erythromycin 500 mg twice daily for four days resulted in ruxolitinib C_max and AUC that were higher by 8% and 27%, respectively, than with ruxolitinib alone.

No dose adjustment is recommended when ruxolitinib is co-administered with mild or moderate CYP3A4 inhibitors (e.g. erythromycin). However, patients should be closely monitored for cytopenias when initiating therapy with a moderate CYP3A4 inhibitor.

No specific dose adjustment is needed in patients with mild or moderate renal impairment.

In patients with severe renal impairment (creatinine clearance less than 30 ml/min) the recommended starting dose based on platelet count for MF patients should be reduced by approximately 50% to be administered twice daily. The recommended starting dose for PV patients with severe renal impairment is 5 mg twice daily. Patients should be carefully monitored with regard to safety and efficacy during ruxolitinib treatment.

There are limited data to determine the best dosing options for patients with end-stage renal disease (ESRD) on haemodialysis. Pharmacokinetic/pharmacodynamic simulations based on available data in this population suggest that the starting dose for MF patients with ESRD on haemodialysis is a single dose of 15-20 mg or two doses of 10 mg given 12 hours apart, to be administered post-dialysis and only on the day of haemodialysis. A single dose of 15 mg is recommended for MF patients with platelet count between 100,000/mm³ and 200,000/mm³. A single dose of 20 mg or two doses of 10 mg given 12 hours apart is recommended for MF patients with platelet count of >200,000/mm³. Subsequent doses (single administration or two doses of 10 mg given 12 hours apart) should be administered only on haemodialysis days following each dialysis session.

The recommended starting dose for PV patients with ESRD on haemodialysis is a single dose of 10 mg or two doses of 5 mg given 12 hours apart, to be administered post-dialysis and only on the day of haemodialysis. These dose recommendations are based on simulations and any dose modification in ESRD should be followed by careful monitoring of safety and efficacy in individual patients. No data is available for dosing patients who are undergoing peritoneal dialysis or continuous venovenous haemofiltration.

Treatment with ruxolitinib has been associated with increases in lipid parameters including total cholesterol, high-density lipoprotein (HDL) cholesterol, low-density lipoprotein (LDL) cholesterol, and triglycerides. Lipid monitoring and treatment of dyslipidaemia according to clinical guidelines is recommended.

Progressive multifocal leukoencephalopathy (PML) has been reported with ruxolitinib treatment. Physicians should be particularly alert to symptoms suggestive of PML that patients may not notice (e.g., cognitive, neurological or psychiatric symptoms or signs). Patients should be monitored for any of these new or worsening symptoms or signs, and if such symptoms/signs occur, referral to a neurologist and appropriate diagnostic measures for PML should be considered. If PML is suspected, further dosing must be suspended until PML has been excluded.

Treatment with ruxolitinib can cause haematological adverse drug reactions, including thrombocytopenia, anaemia and neutropenia. A complete blood count, including a white blood cell count differential, must be performed before initiating therapy with ruxolitinib. Treatment should be discontinued in patients with platelet count less than 50,000/mm³ or absoute neutrophil count less than 500/mm³.

It has been observed that patients with low platelet counts (<200,000/mm³) at the start of therapy are more likely to develop thrombocytopenia during treatment.

Thrombocytopenia is generally reversible and is usually managed by reducing the dose or temporarily withholding ruxolitinib. However, platelet transfusions may be required as clinically indicated.

Patients developing anaemia may require blood transfusions. Dose modifications or interruption for patients developing anaemia may also be considered.

Patients with a haemoglobin level below 10.0 g/dl at the beginning of the treatment have a higher risk of developing a haemoglobin level below 8.0 g/dl during treatment compared to patients with a higher baseline haemoglobin level (79.3% versus 30.1%). More frequent monitoring of haematology parameters and of clinical signs and symptoms of ruxolitinib-related adverse drug reactions is recommended for patients with baseline haemoglobin below 10.0 g/dl.

Neutropenia (absolute neutrophil count <500) was generally reversible and was managed by temporarily withholding ruxolitinib.

Complete blood counts should be monitored as clinically indicated and dose adjusted as required.

Non-melanoma skin cancers (NMSCs), including basal cell, squamous cell, and Merkel cell carcinoma, have been reported in patients treated with ruxolitinib. Most of these patients had histories of extended treatment with hydroxyurea and prior NMSC or pre-malignant skin lesions. A causal relationship to ruxolitinib has not been established. Periodic skin examination is recommended for patients who are at increased risk for skin cancer.

Serious bacterial, mycobacterial, fungal, viral and other opportunistic infections have occurred in patients treated with ruxolitinib. Patients should be assessed for the risk of developing serious infections. Physicians should carefully observe patients receiving ruxolitinib for signs and symptoms of infections and initiate appropriate treatment promptly. Treatment with ruxolitinib should not be started until active serious infections have resolved.

Tuberculosis has been reported in patients receiving ruxolitinib. Before starting treatment, patients should be evaluated for active and inactive (“latent”) tuberculosis, as per local recommendations. This can include medical history, possible previous contact with tuberculosis, and/or appropriate screening such as lung x-ray, tuberculin test and/or interferon-gamma release assay, as applicable. Prescribers are reminded of the risk of false negative tuberculin skin test results, especially in patients who are severely ill or immunocompromised.

Hepatitis B viral load (HBV-DNA titre) increases, with and without associated elevations in alanine aminotransferase and aspartate aminotransferase, have been reported in patients with chronic HBV infections taking ruxolitinib. The effect of ruxolitinib on viral replication in patients with chronic HBV infection is unknown. Patients with chronic HBV infection should be treated and monitored according to clinical guidelines.

Physicians should educate patients about early signs and symptoms of herpes zoster, advising that treatment should be sought as early as possible.

There are no data from the use of ruxolitinib in pregnant women.

Animal studies have shown that ruxolitinib is embryotoxic and foetotoxic. Teratogenicity was not observed in rats or rabbits. However, the exposure margins compared to the highest clinical dose were low and the results are therefore of limited relevance for humans. The potential risk for humans is unknown. As a precautionary measure, the use of ruxolitinib during pregnancy is contraindicated.

Ruxolitinib must not be used during breast-feeding and breast-feeding should therefore be discontinued when treatment is started. It is unknown whether ruxolitinib and/or its metabolites are excreted in human milk. A risk to the breast-fed child cannot be excluded. Available pharmacodynamic/toxicological data in animals have shown excretion of ruxolitinib and its metabolites in milk.

Women of childbearing potential/Contraception

Women of child-bearing potential should use effective contraception during the treatment with ruxolitinib. In case pregnancy should occur during treatment with ruxolitinib, a risk/benefit evaluation must be carried out on an individual basis with careful counselling regarding potential risks to the foetus.

Fertility

There are no human data on the effect of ruxolitinib on fertility. In animal studies, no effect on fertility was observed.

Ruxolitinib has no or negligible sedating effect. However, patients who experience dizziness after the intake of ruxolitinib should refrain from driving or using machines.

Summary of the safety profile

Myelofibrosis

The most frequently reported adverse drug reactions were thrombocytopenia and anaemia.

Haematological adverse drug reactions (any Common Terminology Criteria for Adverse Events [CTCAE] grade) included anaemia (83.8%), thrombocytopenia (80.5%) and neutropenia (20.8%).

Anaemia, thrombocytopenia and neutropenia are dose-related effects.

The three most frequent non-haematological adverse drug reactions were bruising (33.3%), other bleeding (including epistaxis, post-procedural haemorrhage and haematuria) (24.3%) and dizziness (21.9%).

The three most frequent non-haematological laboratory abnormalities identified as adverse reactions were increased alanine aminotransferase (40.7%), increased aspartate aminotransferase (31.5%) and hypertriglyceridaemia (25.2%). In phase 3 clinical studies in MF, neither CTCAE grade 3 or 4 hypertriglyceridaemia or increased aspartate aminotransferase, nor CTCAE grade 4 increased alanine aminotransferase or hypercholesterolaemia were observed.

Discontinuation due to adverse events, regardless of causality, was observed in 30.0% of patients.

Polycythaemia vera

The most frequently reported adverse drug reactions were anaemia and increased alanine aminotransferase.

Haematological adverse reactions (any CTCAE grade) included anaemia (61.8%), thrombocytopenia (25.0%) and neutropenia (5.3%). Anaemia and thrombocytopenia CTCAE grade 3 or 4 were reported in 2.9% and 2.6% of the patients, respectively.

The three most frequent non-haematological adverse reactions were weight gain (20.3%), dizziness (19.4%) and headache (17.9%).

The three most frequent non-haematological laboratory abnormalities (any CTCAE grade) identified as adverse reactions were increased alanine aminotransferase (45.3%), increased aspartate aminotransferase (42.6%), and hypercholesterolaemia (34.7%). No CTCAE grade 4 increased alanine aminotransferase or hypercholesterolaemia, and one CTCAE grade 4 increased aspartate aminotransferase were observed.

Discontinuation due to adverse events, regardless of causality, was observed in 19.4% of patients.

Acute GvHD

The most frequently reported overall adverse drug reactions were thrombocytopenia, anaemia and neutropenia.

Haematological laboratory abnormalities identified as adverse drug reactions included thrombocytopenia (85.2%), anaemia (75.0%) and neutropenia (65.1%). Grade 3 anaemia was reported in 47.7% of patients (grade 4 not applicable per CTCAE v4.03). Grade 3 and 4 thrombocytopenia were reported in 31.3% and 47.7% of patients, respectively.

The three most frequent non-haematological adverse drug reactions were cytomegalovirus (CMV) infection (32.3%), sepsis (25.4%) and urinary tract infections (17.9%).

The three most frequent non-haematological laboratory abnormalities identified as adverse drug reactions were increased alanine aminotransferase (54.9%), increased aspartate aminotransferase (52.3%) and hypercholesterolaemia (49.2%). The majority were of grade 1 and 2.

Discontinuation due to adverse events, regardless of causality, was observed in 29.4% of patients.

Chronic GvHD

The most frequently reported overall adverse drug reactions were anaemia, hypercholesterolemia and increased aspartate aminotransferase.

Haematological laboratory abnormalities identified as adverse drug reactions included anaemia (68.6%), thrombocytopenia (34.4%) and neutropenia (36.2%). Grade 3 anaemia was reported in 14.8% of patients (grade 4 not applicable per CTCAE v4.03). Grade 3 and 4 neutropenia were reported in 9.5% and 6.7% of patients, respectively.

The three most frequent non-haematological adverse drug reactions were hypertension (15.0%), headache (10.2%) and urinary tract infections (9.3%).

The three most frequent non-haematological laboratory abnormalities identified as adverse drug reactions were hypercholesterolaemia (52.3%), increased aspartate aminotransferase (52.2%) and increased alanine aminotransferase (43.1%). The majority were grade 1 and 2.

Discontinuation due to adverse events, regardless of causality, was observed in 18.1% of patients.

Tabulated list of adverse drug reactions from clinical studies

The safety of ruxolitinib in MF patients was evaluated using the long-term follow-up data from two phase 3 studies (COMFORT-I and COMFORT-II) including data from patients initially randomised to ruxolitinib (n=301) and patients who received ruxolitinib after crossing over from control treatments (n=156). The median exposure upon which the adverse drug reaction frequency categories for MF patients are based was 30.5 months (range 0.3 to 68.1 months).

The safety of ruxolitinib in PV patients was evaluated using the long-term follow-up data from two phase 3 studies (RESPONSE, RESPONSE 2) including data from patients initially randomised to ruxolitinib (n=184) and patients who received ruxolitinib after crossing over from control treatments (n=156). The median exposure upon which the adverse drug reaction frequency categories for PV patients are based was 41.7 months (range 0.03 to 59.7 months).

The safety of ruxolitinib in acute GvHD patients was evaluated in the phase 3 study REACH2, including data from patients initially randomised to ruxolitinib (n=152) and patients who received ruxolitinib after crossing over from the best available therapy (BAT) arm (n=49). The median exposure upon which the adverse drug reaction frequency categories were based was 8.9 weeks (range 0.3 to 66.1 weeks).

The safety of ruxolitinib in chronic GvHD patients was evaluated in the phase 3 study REACH3, including data from patients initially randomised to ruxolitinib (n=165) and patients who received ruxolitinib after crossing over from BAT (n=61). The median exposure upon which the adverse drug reaction frequency categories were based was 41.4 weeks (range 0.7 to 127.3 weeks).

In the clinical study programme the severity of adverse drug reactions was assessed based on the CTCAE, defining grade 1=mild, grade 2=moderate, grade 3=severe, grade 4=life-threatening or disabling, grade 5=death.

Adverse drug reactions from clinical studies in MF and PV (Table 1) and in acute and chronic GvHD (Table 2) are listed by MedDRA system organ class. Within each system organ class, the adverse drug reactions are ranked by frequency, with the most frequent reactions first. In addition, the corresponding frequency category for each adverse drug reaction is based on the following convention: very common (≥1/10); common (≥1/100 to <1/10); uncommon (≥1/1,000 to <1/100); rare (≥1/10,000 to <1/1,000); very rare (<1/10,000); not known (cannot be estimated from the available data).

Table 1. Frequency category of adverse drug reactions reported in the phase 3 studies in MF and PV:

^a Frequency is based on new or worsened laboratory abnormalities compared to baseline.
^b Pancytopenia is defined as haemoglobin level <100 g/l, platelet count <100x10⁹/l, and neutrophil count <1.5x10⁹/l (or low white blood cell count of grade 2 if neutrophil count is missing), simultaneously in the same lab assessment
^c Common Terminology Criteria for Adverse Events (CTCAE) version 3.0; grade 1 = mild, grade 2 = moderate, grade 3 = severe, grade 4 = life-threatening
^d These ADRs are discussed in the text.
^e ADR derived from post-marketing experience

Upon discontinuation, MF patients may experience a return of MF symptoms such as fatigue, bone pain, fever, pruritus, night sweats, symptomatic splenomegaly and weight loss. In clinical studies in MF the total symptom score for MF symptoms gradually returned to baseline value within 7 days after dose discontinuation.

Table 2. Frequency category of adverse drug reactions reported in the phase 3 studies in GvHD:

¹ Frequency is based on new or worsened laboratory abnormalities compared to baseline.
² Pancytopenia is defined as haemoglobin level <100 g/l, platelet count <100 × 10⁹/l, and neutrophil count <1.5 × 10⁹/l (or low white blood cell count of grade 2 if neutrophil count is missing), simultaneously in the same laboratory assessment.
³ CTCAE Version 4.03.
⁴ Grade ≥3 sepsis includes 20 (10%) grade 5 events.
⁵ Not applicable: no cases reported

Description of selected adverse drug reactions

Anaemia

In phase 3 clinical studies in MF, median time to onset of first CTCAE grade 2 or higher anaemia was 1.5 months. One patient (0.3%) discontinued treatment because of anaemia.

In patients receiving ruxolitinib mean decreases in haemoglobin reached a nadir of approximately 10 g/litre below baseline after 8 to 12 weeks of therapy and then gradually recovered to reach a new steady state that was approximately 5 g/litre below baseline. This pattern was observed in patients regardless of whether they had received transfusion during therapy.

In the randomised, placebo-controlled study COMFORT-I 60.6% of ruxolitinib-treated MF patients and 37.7% of placebo-treated MF patients received red blood cell transfusions during randomised treatment. In the COMFORT-II study the rate of packed red blood cell transfusions was 53.4% in the ruxolitinib arm and 41.1% in the best available therapy arm.

In the randomised period of the pivotal studies, anaemia was less frequent in PV patients than in MF patients (40.8% versus 82.4%). In the PV population, the CTCAE grade 3 and 4 events were reported in 2.7%, while in the MF patients the frequency was 42.56%.

In the phase 3 acute and chronic GvHD studies, anaemia CTCAE grade 3 was reported in 47.7% and 14.8% of patients, respectively.

Thrombocytopenia

In the phase 3 clinical studies in MF, in patients who developed grade 3 or 4 thrombocytopenia, the median time to onset was approximately 8 weeks. Thrombocytopenia was generally reversible with dose reduction or dose interruption. The median time to recovery of platelet counts above 50,000/mm³ was 14 days. During the randomised period, platelet transfusions were administered to 4.7% of patients receiving ruxolitinib and to 4.0% of patients receiving control regimens. Discontinuation of treatment because of thrombocytopenia occurred in 0.7% of patients receiving ruxolitinib and 0.9% of patients receiving control regimens. Patients with a platelet count of 100,000/mm³ to 200,000/mm³ before starting ruxolitinib had a higher frequency of grade 3 or 4 thrombocytopenia compared to patients with platelet count >200,000/mm³ (64.2% versus 38.5%).

In the randomised period of the pivotal studies, the rate of patients experiencing thrombocytopenia was lower in PV (16.8%) patients compared to MF (69.8%) patients. The frequency of severe (i.e. CTCAE grade 3 and 4) thrombocytopenia was lower in PV (2.7%) than in MF (11.6%) patients.

In the phase 3 acute GvHD study, grade 3 and 4 thrombocytopenia was observed in 31.3% and 47.7% of patients, respectively. In the phase 3 chronic GvHD study, grade 3 and 4 thrombocytopenia was lower (5.9% and 10.7%) than in acute GvHD.

Neutropenia

In the phase 3 clinical studies in MF, in patients who developed grade 3 or 4 neutropenia, the median time to onset was 12 weeks. During the randomised period, dose holding or reductions due to neutropenia were reported in 1.0% of patients, and 0.3% of patients discontinued treatment because of neutropenia.

In the randomised period of the phase 3 studies in PV patients, neutropenia was reported in 1.6% of patients exposed to ruxolitinib compared to 7% in reference treatments. In the ruxolitinib arm one patient developed CTCAE grade 4 neutropenia. An extended follow-up of patients treated with ruxolitinib showed 2 patients reporting CTCAE grade 4 neutropenia.

In the phase 3 acute GvHD study, grade 3 and 4 neutropenia was observed in 17.9% and 20.6% of patients, respectively. In the phase 3 chronic GvHD study grade 3 and 4 neutropenia was lower (9.5% and 6.7%) than in acute GvHD.

Bleeding

In the phase 3 pivotal studies in MF bleeding events (including intracranial and gastrointestinal, bruising and other bleeding events) were reported in 32.6% of patients exposed to ruxolitinib and 23.2% of patients exposed to the reference treatments (placebo or best available therapy). The frequency of grade 3-4 events was similar for patients treated with ruxolitinib or reference treatments (4.7% versus 3.1%). Most of the patients with bleeding events during the treatment reported bruising (65.3%). Bruising events were more frequently reported in patients taking ruxolitinib compared with the reference treatments (21.3% versus 11.6%). Intracranial bleeding was reported in 1% of patients exposed to ruxolitinib and 0.9% exposed to reference treatments. Gastrointestinal bleeding was reported in 5.0% of patients exposed to ruxolitinib compared to 3.1% exposed to reference treatments. Other bleeding events (including events such as epistaxis, post-procedural haemorrhage and haematuria) were reported in 13.3% of patients treated with ruxolitinib and 10.3% treated with reference treatments.

During the long-term follow-up of phase 3 clinical studies in MF, the cumulative frequency of bleeding events increased proportionally to the increase in the follow-up time. Bruising events were the most frequently reported bleeding events (33.3%). Intracranial and gastrointestinal bleeding events were reported in 1.3% and 10.1% of patients respectively.

In the comparative period of phase 3 studies in PV patients, bleeding events (including intracranial and gastrointestinal, bruising and other bleeding events) were reported in 16.8% of patients treated with ruxolitinib, 15.3% of patients receiving best available therapy in RESPONSE study and 12.0% of patients receiving best available therapy in RESPONSE 2 study. Bruising was reported in 10.3% of patients treated with ruxolitinib, 8.1% of patients receiving best available therapy in RESPONSE study and 2.7% of patients receiving best available therapy in RESPONSE 2 study. No intracranial bleeding or gastrointestinal haemorrhage events were reported in patients receiving ruxolitinib. One patient treated with ruxolitinib experienced a grade 3 bleeding event (post-procedural bleeding); no grade 4 bleeding was reported. Other bleeding events (including events such as epistaxis, post-procedural haemorrhage, gingival bleeding) were reported in 8.7% of patients treated with ruxolitinib, 6.3% of patients treated with best available therapy in RESPONSE study and 6.7% of patients treated with best available therapy in RESPONSE 2 study.

During the long-term follow-up of phase 3 studies in PV, the cumulative frequency of bleeding events increased proportionally to the increase in the follow-up time. Bruising events were the most frequently reported bleeding events (17.4%). Intracranial and gastrointestinal bleeding events were reported in 0.3% and 3.5% of patients respectively.

In the comparative period of the phase 3 acute GvHD study, bleeding events were reported in 25.0% and 22.0% of patients in the ruxolitinib and BAT arms respectively. The sub-groups of bleeding events were generally similar between treatment arms: bruising events (5.9% in ruxolitinib vs. 6.7% in BAT arm), gastrointestinal events (9.2% vs. 6.7%) and other haemorrhage events (13.2% vs. 10.7%). Intracranial bleeding events were reported in 0.7% of patients in the BAT arm and in no patients in the ruxolitinib arm.

In the comparative period of the phase 3 chronic GvHD study, bleeding events were reported in 11.5% and 14.6% of patients in the ruxolitinib and BAT arms respectively. The sub-groups of bleeding events were generally similar between treatment arms: bruising events (4.2% in ruxolitinib vs. 2.5% in BAT arm), gastrointestinal events (1.2% vs. 3.2%) and other haemorrhage events (6.7% vs. 10.1%). No intracranial bleeding events were reported in either treatment arm.

Infections

In the phase 3 pivotal studies in MF, grade 3 or 4 urinary tract infection was reported in 1.0% of patients, herpes zoster in 4.3% and tuberculosis in 1.0%. In phase 3 clinical studies sepsis was reported in 3.0% of patients. An extended follow-up of patients treated with ruxolitinib showed no trends towards an increase in the rate of sepsis over time.

In the randomised period of the phase 3 studies in PV patients, one (0.5%) CTCAE grade 3 and no grade 4 urinary tract infection was reported. The rate of herpes zoster was similar in PV (4.3%) patients and MF (4.0%) patients. There was one report of CTCAE grade 3 post-herpetic neuralgia amongst the PV patients. Pneumonia was reported in 0.5% of patients treated with ruxolitinib compared to 1.6% of patients in reference treatments. No patients in the ruxolitinib arm reported sepsis or tuberculosis.

During long-term follow-up of phase 3 studies in PV, frequently reported infections were urinary tract infections (11.8%), herpes zoster (14.7%) and pneumonia (7.1%). Sepsis was reported in 0.6% of patients. No patients reported tuberculosis in long-term follow-up.

In the phase 3 acute GvHD study, during the comparative period, urinary tract infections were reported in 9.9% (grade ≥3, 3.3%) of patients in the ruxolitinib arm compared to 10.7% (grade ≥3, 6.0%) in the BAT arm. CMV infections were reported in 28.3% (grade ≥3, 9.3%) of patients in the ruxolitinib arm compared to 24.0% (grade ≥3, 10.0%) in the BAT arm. Sepsis events were reported in 12.5% (grade ≥3, 11.1%) of patients in the ruxolitinib arm compared to 8.7% (grade ≥3, 6.0%) in the BAT arm. BK virus infection was reported only in the ruxolitinib arm in 3 patients with one grade 3 event. During extended follow-up of patients treated with ruxolitinib, urinary tract infections were reported in 17.9% (grade ≥3, 6.5%) of patients and CMV infections were reported in 32.3% (grade ≥3, 11.4%) of patients. CMV infection with organ involvement was seen in very few patients; CMV colitis, CMV enteritis and CMV gastrointestinal infection of any grade were reported in four, two and one patients, respectively. Sepsis events, including septic shock, of any grade were reported in 25.4% (grade ≥3, 21.9%) of patients.

In the phase 3 chronic GvHD study, during the comparative period, urinary tract infections were reported in 8.5% (grade ≥3, 1.2%) of patients in the ruxolitinib arm compared to 6.3% (grade ≥3, 1.3%) in the BAT arm. BK virus infection was reported in 5.5% (grade ≥3, 0.6%) of patients in the ruxolitinib arm compared to 1.3% in the BAT arm. CMV infections were reported in 9.1% (grade ≥3, 1.8%) of patients in the ruxolitinib arm compared to 10.8% (grade ≥3, 1.9%) in the BAT arm. Sepsis events were reported in 2.4% (grade ≥3, 2.4%) of patients in the ruxolitinib arm compared to 6.3% (grade ≥3, 5.7%) in the BAT arm. During extended follow-up of patients treated with ruxolitinib, urinary tract infections and BK virus infections were reported in 9.3% (grade ≥3, 1.3%) and 4.9% (grade ≥3, 0.4%) of patients, respectively. CMV infections and sepsis events were reported in 8.8% (grade ≥3, 1.3%) and 3.5% (grade ≥3, 3.5%) of patients, respectively.

Elevated lipase

In the randomised period of the RESPONSE study, the worsening of lipase values was higher in the ruxolitinib arm compared to the control arm, mainly due to the differences among grade 1 elevations (18.2% vs 8.1%). Grade ≥2 elevations were similar between treatment arms. In RESPONSE 2, the frequencies were comparable between the ruxolitinib and the control arm (10.8% vs 8%). During longterm follow-up of phase 3 PV studies, 7.4% and 0.9% of patients reported grade 3 and grade 4 elevation of lipase values. No concurrent signs and symptoms of pancreatitis with elevated lipase values were reported in these patients.

In phase 3 studies in MF, high lipase values were reported in 18.7% and 19.3% of patients in the ruxolitinib arms compared to 16.6% and 14.0% in the control arms in COMFORT-I and COMFORT-II studies, respectively. In patients with elevated lipase values, no concurrent signs and symptoms of pancreatitis were reported.

In the comparative period of the phase 3 acute GvHD study, new or worsened lipase values were reported in 19.7% of patients in the ruxolitinib arm compared to 12.5% in the BAT arm; corresponding grade 3 (3.1% vs 5.1%) and grade 4 (0% vs 0.8%) increases were similar. During extended follow-up of patients treated with ruxolitinib, increased lipase values were reported in 32.2% of patients; grade 3 and 4 were reported in 8.7% and 2.2% of patients respectively.

In the comparative period of the phase 3 chronic GvHD study, new or worsened lipase values were reported in 32.1% of patients in the ruxolitinib arm compared to 23.5% in the BAT arm; corresponding grade 3 (10.6% vs 6.2%) and grade 4 (0.6% vs 0%) increases were similar. During extended follow-up of patients treated with ruxolitinib, increased lipase values were reported in 35.9% of patients; grade 3 and 4 were observed in 9.5% and 0.4% of patients, respectively.

Increased systolic blood pressure

In the phase 3 pivotal clinical studies in MF an increase in systolic blood pressure of 20 mmHg or more from baseline was recorded in 31.5% of patients on at least one visit compared with 19.5% of the control-treated patients. In COMFORT-I (MF patients) the mean increase from baseline in systolic BP was 0-2 mmHg on ruxolitinib versus a decrease of 2-5 mmHg in the placebo arm. In COMFORT-II mean values showed little difference between the ruxolitinib-treated and the control-treated MF patients.

In the randomised period of the pivotal study in PV patients, the mean systolic blood pressure increased by 0.65 mmHg in the ruxolitinib arm versus a decrease of 2 mmHg in the BAT arm.

Paediatric patients

A total of 20 patients aged 12 to <18 years with GvHD were analysed for safety: 9 patients (5 in the ruxolitinib arm and 4 in the BAT arm) in the study REACH2 and 11 patients (4 in the ruxolitinib arm and 7 in the BAT arm) in the study REACH3. Based on the similar exposure observed in adolescents and adults, the safety of ruxolitinib at the recommended dose of 10 mg twice daily is similar in frequency and severity.

Elderly

A total of 29 patients in study REACH2 and 25 patients in REACH3 aged >65 years and treated with ruxolitinib were analysed for safety. Overall, no new safety concerns were identified and the safety profile in patients >65 years old is generally consistent with that of patients aged 18-65 years old.