Imetelstat for patients with highly transfusion-dependent MDS

Myelodysplastic syndromes (MDS) are a group of disorders in which hematopoiesis is ineffective, resulting in cytopenias, and clonal hematopoiesis that can eventually progress to acute myeloid leukemia. Telomerase activity and human telomerase reverse transcription expression (hTERT) are often increased in clonal hematopoietic cells of patients with MDS. These factors along with others, such as red blood cell (RBC) transfusion dependence, confer poor prognosis.

The initial treatment option for patients with lower-risk MDS is the use of erythropoiesis-stimulating agents (ESAs) with/without other hematopoietic growth factors and transfusions. However, only approximately 40% of patients will achieve hematologic improvement for a median duration of 2 years, and resistance/relapse to ESAs often occur. Therefore, there is an unmet clinical need for patients who are ESA-refractory.¹

Imetelstat is a first-in-class telomerase inhibitor. It is composed of a 13-mer oligonucleotide that specifically targets the RNA template of telomerase, and thus blocks its activity. The phase II results of the IMerge (NCT02598661) global phase II/III study of imetelstat in patients with RBC transfusion dependent, ESA-relapsed/refractory, lower-risk MDS were recently published in the Journal of Clinical Oncology, by David Steensma and colleagues.¹ Latest updates were presented by Uwe Platzbecker at the 62nd American Society of Hematology (ASH) Annual Meeting and Exposition and are summarized below.²   

Study design and patient characteristics

• Open label, single-arm, multicenter study of imetelstat in 57 adult patients with RBC transfusion dependent, ESA-relapsed/refractory, lower-risk MDS
  • As an initial analysis revealed higher hematological response rates for patients who were non-del(5q) and hypomethylating agent (HMA)/lenalidomide naïve, these patients were analyzed as a subpopulation (n = 38) 
• 7.5 mg/kg imetelstat was administered as a 2-hour intravenous infusion every 4 weeks, until disease progression, unacceptable toxicity, withdrawal of consent or lack of response 
• Primary endpoint: 8-week RBC transfusion independence (TI) rate
• Secondary endpoints: 24-week RBC TI rate; time to TI onset; duration of TI; hematologic improvement-erythroid response (HI-E) rate; MDS response (complete response [CR], marrow CR [mCR], or partial response, per 2006 International Working Group (IWG) criteria; and safety
• Baseline characteristics can be seen in Table 1
  • Baseline transfusion burden was high, at a median of 7 units/8 weeks
  • Eastern Cooperative Oncology Group (ECOG), International Prognostic Scoring System (IPSS), and IPSS-revised risk were comparable in the subset population
  • The subset population had lower serum erythropoietin (sEPO) despite similar prior ESA use

Table 1. Notable baseline patient characteristics¹

ECOG PS, Eastern Cooperative Oncology Group performance status; ESA, erythropoiesis-stimulating agent; RBC, red blood cell; sEPO, serum erythropoietin.
Characteristic	Overall population (N = 57)	Subset (n = 38)
Median age, years (range)	71.0 (46–83)	71.5 (46–83)
male/female, %	56/44	66/34
ECOG PS 0–1, %	91	89
Median RBC transfusion burden, units/8-weeks (range)	7 (4–14)	8 (4–14)
> 4 units/8 weeks at baseline, %	93	92
Prior ESA use, %	90	89
sEPO > 500 mU/mL, %	40	32

Results

Efficacy

For the overall population:¹

• Clinical cutoff: April 30, 2019
• Median follow-up: 16.4 months (range, 3.9–37.5 months) overall population
• Overall median treatment duration was 8.2 months (range, 0.02–37.5 months), with a median of eight treatment cycles

For the subset population:²

• Clinical cutoff: February 4, 2020
• Median follow-up: 24 months (n =38). Of the total patients, 11 (29%) dropped out of the study, 8 (21%) due to death 3 (8%) withdrew themselves
• Median treatment duration was 8.5 months with a median of nine treatment cycles

Efficacy outcomes are summarized in Table 2.

• 8-week RBC TI rate was 37% and 42% in the overall and subset population, respectively
  • This did not differ based on baseline RBC transfusion burden, presence of ringed sideroblasts, or baseline serum erythropoietin levels
• 24-week RBC TI rate was 23% and 32% in the overall and subset population, respectively
• The median time to onset of TI was similar in the subset population as the overall population (8.3 months) but the duration of TI was longer in the subset population (88 weeks vs 65 weeks)
• A clinically meaningful major response (16-week RBC TI, per 2018 IWG guidelines), was reached by 28% of the overall population and was slightly higher in the subset population (37%)

Table 2. Efficacy outcomes

HI-E, hematologic improvement-erythroid; Hgb, hemoglobin; IWG, International Working Group; TI, transfusion independence.
Outcome	Overall population1 (n = 57)	Subset2 (n = 38)
8-week TI, %	37	42
Median time to onset, weeks (range)	8.3 (0.1–100.6)	8.3 (0.1–40.7)
Median duration of TI, weeks (range)	65 (17.0–140.9)	88.0 (23.1–140.9)
24-week TI, %	23	32
HI-E per 2006 IWG, %	65	68
≥ 1.5 g/dL increase in Hgb lasting ≥ 8 weeks, %	26	34
Transfusion reduction by ≥ 4 units/8 weeks	65	68
Response per 2018 IWG, %
Major response: 16-week TI	28	37
Major response: 8-week TI	37	42
Minor response	49	55

• Overall response rate was 19% for the overall population
  • 9% of patients achieved CR
  • 11% of patients achieved mCR
• Overall response rate was 24% for the subset population
  • 11% of patients achieved CR
  • 13% of patients achieved mCR
  • Over half the patients (68%) showed hematologic improvement with a median duration of 92.7 weeks (95% CI, 37.1–149.4)
• Of the patients that achieved CR or mCR, 46% had complete resolution of dysplasia
  • Of the patients that had ringed sideroblasts on screening, 50% had complete disappearance of ringed sideroblasts and the other 50% had a ≥ 30% reduction in these cells.

Mutational analysis¹

Of the 13 patients that had pre- and posttreatment mutational analysis, 11 had baseline SF3B1 mutations. A reduction in variant allele frequency (VAF) was seen in 10 patients and the greater the reduction, the longer the TI duration. Patients that achieved a ≥ 50% reduction in VAF maintained TI for > 18 months. This was a significant correlation (Pearson correlation coefficient, r = 0.646; p = 0.032).

Biomarkers

• In the overall population, a ≥ 50% reduction in telomerase activity and hTERT expression from baseline was observed in 38% and 59% of patients, respectively
  • In the subset population this was 23.1% and 54.3% of patients, respectively
• In the overall population, a higher proportion of patients who had ≥ 50% reduction in hTERT expression achieved 8-week and 24-week TI compared with those that did not achieve TI (Table 3)
  • This effect was significantly higher in the subset population

Table 3. Percentage of patients with ≥ 50% reduction in hTERT expression in those with and without TI

hTERT, human telomerase reverse transcription; TI, transfusion independence.
% of patients with ≥ 50% reduction in hTERT expression	Overall population1	p value	Subset2	p value
With 8-week TI	75	0.083	80	0.016
Without 8-week TI	48		35
With 24-week TI	85	0.048	91.7	0.002
Without 24-week TI	50		34.8

Safety¹

• The frequency and type of treatment-emergent adverse events (TEAEs) were similar in the subset population and overall population
• The most frequent Grade ≥ 3 TEAEs (Table 4) were neutropenia (60%), thrombocytopenia (54%), and anemia (19%):
  • Only one patient had anemia, which was assessed as related to imetelstat use
  • 90% of Grade 3/4 neutropenia events were reversible within 4 weeks
  • 88% of Grade 3/4 thrombocytopenia events were reversible within 4 weeks

Table 4. Grade ≥ 3 TEAEs with incidence > 5%¹

ALT, alanine aminotransferase; AST, aspartate aminotransferase; TEAE, treatment-emergent adverse event.
Grade ≥ 3 TEAE	Overall population (N = 57)	Subset (n = 38)
Hematologic, %
Thrombocytopenia	54	61
Neutropenia	60	55
Anemia	19	21
Nonhematologic, %
Back pain	5	5
ALT increase	5	5
AST increase	5	8
Bronchitis	5	8

Conclusions

Imetelstat treatment for patients with RBC transfusion-dependent, ESA-relapsed/refractory, lower-risk MDS, demonstrated a predictable safety profile and induced durable and clinically meaningful responses in terms of 8-week and 24-week TI. The effects were greater in the subset of patients that were non-del(5q) and HMA/lenalidomide naïve. There was also evidence of disease-modifying activity of imetelstat by reduction of cytogenetically abnormal clones and mutational allele burden; however, this was only tested in a small number of patients.

The ongoing phase III portion of the trial will continue to assess the efficacy and safety of imetelstat in a larger cohort of patients in a placebo-controlled, double-blind manner.³