Five-year follow-up of the RESPONSE phase III trial: ruxolitinib for polycythemia vera

Polycythemia vera (PV) is a rare myeloproliferative neoplasm (MPN) associated with a high risk of vascular events and substantial disease burden that impair the quality of life of patients.¹ Hydroxyurea has been the standard of care for PV for a long time, however, a proportion of patients with PV become resistant or are intolerant to hydroxyurea.^2,3 There are only few promising first-line treatment alternatives which are mostly consisting of peginterferon alpha compounds (read more here). Nevertheless, together with the fact that in more than 95% of patients the underlying PV etiology is a mutation in the Janus kinase 2 (JAK2; V617P) gene, the JAK inhibitor, ruxolitinib, evolved as a very promising therapeutic target.²

In fact, the primary results from the RESPONSE phase III trial led to the approval of ruxolitinib by the U.S. Food and Drug Administration and European Medicines Agency for PV patients resistant or intolerant to hydroxyurea.³ The long-term efficacy and safety of ruxolitinib in this patient subset was investigated in a 5-year follow-up study of the RESPONSE phase III trial by Jean-Jacques Kiladjian and colleagues.² The results of the study were published in the Lancet Haematology and are summarized below.

Study design

• Five-year follow up of the multicenter, international, randomized, open-label, phase III of RESPONSE trial
• Patients were ≥ 18 years old with PV and resistant or intolerant to hydroxyurea
• Patients were excluded in they received peginterferon alpha compounds or phosphorus 32 within 5 weeks of study initiation
• N = 222 patients were randomly 1:1 assigned to either:
  • Ruxolitinib (n = 110): starting at 10 mg orally, twice per day, or;
  • Single agent of best available therapy at physician’s discretion (n = 112)
• The primary composite endpoint was the number of patients achieving both hematocrit control without phlebotomy and ≥ 35% reduction in spleen size at week 32
• The main secondary endpoint was the number of patients achieving complete hematological response

Results

• The key efficacy data from the primary analysis of RESPONSE, as previously reported³, are shown in Table 1

Table 1. Key efficacy and safety results at week 32 of the RESPONSE phase III trial^2,3

*Primary composite endpoint, proportion of patients achieving both hematocrit control without phlebotomy and ≥ 35% reduction in spleen size at week 32 In bold, statistically significant values (p < 0.05)
Ruxolitinib                                                 (n = 110)		Best available therapy (n = 112)	p value
Patients achieving the primary endpoint*	20.9%	0.9%	< 0.001
Patients achieving hematocrit control	60.0%	19.6%	—
Patients with ≥ 35% spleen size reduction	38.2%	0.9%	—
Complete hematological remission	23.6%	8.9%	0.003

•  From the original cohort, 72 (65%) patients from the ruxolitinib group and 64 (57%) patients from the best available therapy arm completed a 5-year treatment (256 weeks)
• The key efficacy results after 5 years of treatment, at week 256, are shown in Table 2
• From week 80 to week 256, 83% of patients in the ruxolitinib group did not require any phlebotomies
• Similarly, from the 79 patients that crossed over from best available therapy to ruxolitinib, 69 (87%) did not require any phlebotomies from week 80 to week 256
• Overall, most patients in the ruxolitinib arm (89%) and those who crossed over (86%) had a decrease in spleen size during the study, while only 49% of those in the best available therapy arm showed similar effects

Table 2. Key efficacy results at week 256 of the 5-year follow-up of the RESPONSE phase III trial²

	Ruxolitinib (n=72)
Progressing after achieving the primary endpoint at week 32	24%
Median duration of primary response	Not reached
Progressing after achieving hematocrit control at week 32	24%
Maintaining hematocrit control (week 32 to week 224)	73% (95% CI, 60–83)
Progressing after achieving complete hematological remission at week 32	38%
Maintaining complete hematological control (week 32 to week 224)	55% (95% CI, 32–73)
Maintaining at least a 35% reduction in spleen size (week 32 to week 224)	72% (95% CI, 34–91)
Progressing after achieving complete clinicohematological response at week 32	30%
Maintaining complete clinicohematological response (week 32 to week 224)	67% (95% CI, 54–77)
Median duration of clinicohematological response	Not reached

Safety

• As patients were allowed to crossover to the ruxolitinib group, safety was also considered by the authors as differences in exposure duration between the groups
  • No increase in exposure-adjusted rates of adverse events (AEs) were observed
  • No new or unexpected AEs were reported
• The most common ruxolitinib-related hematological AE was anemia (rate per 100 patient years, ruxolitinib: 8.9 versus best available therapy: 5.4), with most events being of mild-to-moderate severity while thrombocytopenia was higher in the best available therapy arm (ruxolitinib: 4.4 versus BAT: 16.3; Table 3)
• The incidence of non-hematological AEs was lower in the ruxolitinib arm than in the best available therapy group (Table 3)
• The most common non-hematological AEs were infections, which were lower in the ruxolitinib group (18.9 versus 59.8) except for herpes zoster infections (Table 3)
• Similarly, the rates of serious AEs were lower in the ruxolitinib group (10.3) than the best available therapy arm (13.6)

Table 3. Most common all-grade hematological and non-hematological AE rates from the 5-year follow-up of the RESPONSE phase III trial²

AEs, adverse events *Exposure-adjusted rates ≥ 5 per 100 patient-years
	Ruxolitinib (n = 110)		Best available therapy (n = 111)	Crossover group (n = 98)
Hematological AE rates* (all grades)
Anemia		8.9	5.4	8.8
Thrombocytopenia		4.4	16.3	1.2
Non-hematological AE rates* (all grades)
Pruritus		7.0	32.6	6.1
Diarrhea		7.0	12.2	3.6
Weight gain		6.1	1.4	4.2
Headache		5.8	28.5	5.2
Arthralgia		5.6	10.9	3.3
All infections Herpes zoster		18.9 4.7	59.8 0.0	19.1 3.9

•  The rate of thromboembolic events was lower in patients receiving ruxolitinib (1.2) or belonging to the crossover population (2.7) versus those in the best available group (8.2)
• Details of the most common thromboembolic events occurring during the 5-year follow-up of the RESPONSE trial are shown in Table 4 

Table 4. Most common thromboembolic event rates from the 5-year follow-up of the RESPONSE phase III trial²

*Exposure-adjusted rates ≥ 5 per 100 patient-years $Other thromboembolic events include: for best available therapy, splenic infarction; and for the crossover group, bone infarction, disseminated intravascular coagulation and coronary artery occlusion
		Ruxolitinib (n = 110)	Best available therapy (n = 111)	Crossover group (n = 98)
Thromboembolic event rates* (all grades)
All events		1.2	8.2	2.7
Cerebral infarction		0.2	0	0
Ischemic stroke		0.2	0	0.3
Pulmonary embolism		0.2	1.4	0
Portal vein thrombosis		0.2	0	0
Retinal vascular thrombosis		0.2	0	0
Deep vein thrombosis		0	2.7	0.3
Myocardial infarction		0	0	0.6
Transient ischemic attack		0	0	0.6
Thrombophlebitis		0	1.4	0.3
Thrombosis		0	1.4	0.3
Other$		0	1.4	0.9

 

• Two deaths occurred in the ruxolitinib arm, of which one was deemed treatment-related (gastric adenocarcinoma)

Conclusions

The results from the 5-year follow-up of the RESPONSE phase III trial indicate that ruxolitinib continues to be superior to best available therapy (including interferons), with a substantial proportion of patients maintaining deep durable responses. Moreover, in the long term, ruxolitinib led to fewer thromboembolic events and lower rates of serious AEs, when compared to best available treatment. All these beneficial effects of ruxolitinib were also observed in the crossover patients despite having a shorter exposure to the treatment. In summary, these results provide strong evidence that ruxolitinib is a safe and effective long-term regimen for patients with PV who cannot be treated with hydroxyurea.