Treatment of older or unfit patients with AML

At the 1^st NCRI AML academy meeting, held on the 19^th and 20^th September 2019, Professor Courtney DiNardo  from the MD Anderson Cancer Center, Houston, US discussed the optimal treatment strategy for elderly or frail patients with acute myeloid leukemia (AML). She began by introducing AML as being particularly prevalent in older people, with a median age at diagnosis of 68 years, and by highlighting that older age at diagnosis correlates with poor survival. Two case studies were then presented to encourage the audience to consider treatment of different cases.

Patient one

• 74 year old male
• Fatigue and shortness of breath on exertion
• History of coronary artery disease (had coronary artery bypass graft), atrial fibrillation, ejection fraction (EF) 50% and type 2 diabetes mellitus
• White blood cell (WBC) count 1,500/ml, Hemoglobin 7.5g/dl, platelets count 85,000/ml, 27% circulating blasts
• Eastern Cooperative Oncology Group (ECOG) performance status (PS): one
• Patient interested in leukemia directed therapy
• Diagnosis: AML with 32% CD33+ blasts, diploid cytogenetics, and mutations of NPM1, IDH1, and FLT3-TKD
• Audience vote:
  • 42% of audience voted for the prioritization of intensive induction chemotherapy
  • 45% voted for non-intensive induction chemotherapy
  • 13% voted for ‘not sure/need more information’

Patient 2

• 68 year old female
• Former smoker, has osteoarthritis
• WBC count 2,700/ml, Hemoglobin 7.8g/dl, platelets count 45,000/ml, no circulating blasts
• ECOG PS one
• Patient Interested in leukemia directed therapy
• Diagnosis: AML with 45% CD33+ blasts in the bone marrow (BM), and myelodysplasia (MDS)-related changes, complex karyotype with del(5q), del(20q) and -7, and mutations of RUNX1 and SRSF2
• Audience vote: 
  • 44% of the audience voted for the prioritization of intensive induction chemotherapy
  • 46% voted for non-intensive induction chemotherapy
  • 10% voted for ‘not sure/need more information’

Courtney DiNardo proceeded to talk about the principle of AML therapy, and how historically there were two main cohorts of patients with young, fit patients tending to be treated with intensive chemotherapy followed by hematopoietic stem cell transplantation with a curative intent, and older, less fit patients being given less intensive options with the aim to provide the best supportive palliative care. She stated that treatment is less binary now, owing to advances in AML treatment, where long term relapse-free survival is also seen in some older, less fit patients.

Data from a French study¹ demonstrated that an intensive chemotherapy approach in patients with core binding factor (CBF) AML who were ≥60 years old achieved a complete response (CR) rate of 88%, a median overall survival (OS) of two years and a five year OS of 31%. This study highlights that there is no issue getting older and less fit patients into remission, but long-term they don’t do so well. A further study from the UK (AML2016)² demonstrated that in older patients, receiving intensive therapy did better (CR/incomplete CR (CRi) of 70%) than those receiving non-intensive therapy achieved a CR/Cri of 28%initially, but there was only a small difference in terms of OS at three years (15 –25% vs 5 –15%).

The definition of an unfit patient typically includes:

• Aged 75 or over
• ECOG PS two or higher
• Has underlying organ dysfunction such as kidney failure, liver disease or cardiomyopathy

It is currently debated whether the definition should also include:

• Has received prior treatment for hematological disorders
• Has an unfavorable or high-risk karyotype

Professor DiNardo then presented data from a study of low-dose cytarabine (LDAC) vs hydroxyurea (HU; best supportive care) in patients with AML who were unfit for more intensive treatments,³ which demonstrated improved OS in the LDAC treated group vs the HU group (~25% vs <10% OS at one year). Two further studies of hypomethylating agents (HMAs) demonstrated improvement in survival rates in patients with newly diagnosed AML who were aged 65 years and older with an improved one year OS rate with azacitidine as monotherapy vs best supportive care (46.5% vs 34.2%) and older ⁴ and a higher median OS when treating with decitabine (7.7 vs 5.0 months) compared with best supportive care.⁵

Recently, studies in which a new drug, venetoclax, which was used in combination with LDAC or HMAs, demonstrating improved outcomes of LDAC + venetoclax (CR/CRi 54%; N=82)⁶ over LDAC alone (CR/CRi 11%; N=243)⁵ , and of azacitidine + venetoclax (CR/CRi 71%, median OS 16.9 months; N=84)⁸ over azacitidine alone (CR/CRi 28%, median OS 10.4 months; N=241)⁴ which has led to the accelerated approval of these combination therapies by the FDA.  Response rates by patient subgroups are detailed in Table 1⁸ and treatment-emergent adverse events (TEAEs) in Table 2⁷ for these combination treatment studies. Professor DiNardo showed that certain molecular subtypes of AML respond better to venetoclax combinations than others. For example, NPM1 mutant AML has high CR/CRi rates of 100% and 90% with LDAC + venetoclax or HMA + venetoclax respectively.⁹ In contrast, FLT3 mutant AML responds more favorably to HMA + venetoclax than to LDAC + venetoclax (CR/CRi 71% vs 44%).^6,7  Molecular subtype of AML also has an impact on the duration of response with NPM1 and spliceosome mutated AML variants having the best 30 month OS (~70% and ~40% respectively). Initial results of ten day decitabine treatment (DAC10) + venetoclax prior to stem-cell transplant (SCT) are promising, with an estimated 1-month OS of 100% (n=11) in patients with newly diagnosed AML, and an estimated 1-month OS of 80% (n=9) in patients with relapsed/refractory AML at ~11 months¹⁰. With these encouraging results, venetoclax, HMAs and LDAC have established themselves as a safe and efficacious backbones for the addition of other targeted therapies.

Table 1. Response rates (CR/CRi) by patient subgroups

 

Cytogenetic risk

AML type

Genetic mutation

 

Intermediate

Poor

De novo

Secondary

TP53

IDH1/2

FLT3

NPM1

Azacitidine + venetoclax

38/50 (76%)

22/33 (67%)

48/63 (76%)

12/21 (57%)

13/20 (65%)

18/20 (90%)

8/11 (73%)

11/14 (79%)

Decitabine + venetoclax

11/16 (69%)

12/15 (80%)

16/22 (73%)

7/9 (78%)

6/7 (86%)

5/5 (100%)

1/3 (33%)

3/3 (100%)

Table 2. Treatment-emergent adverse events (TEAE)⁸

TEAE (≥ 25%)

n (%) of Patients

Any Grade (n=145)

Grade 3/4

Nausea

88 (61)

2 (1)

Diarrhea

76 (52)

7 (5)

Constipation

70 (48)

7 (1)

Febrile Neutropenia

63 (43)

63 (43)

Fatigue

54 (37)

8 (6)

Decreased appetite

48 (33)

3 (2)

Leukopenia

45 (31)

45 (31)

Thrombocytopenia

42 (29)

35 (24)

Anemia

40 (28)

36 (25)

Edema

41 (28)

0

In patients with IDH1 mutated AML, treatment with ivosidenib resulted in an OS of 12.6 months¹¹, and when administered in combination with azacitidine outcome is markedly improved with CR rates of 60%, and a 12-month OS of 82%.¹²

Data on older and unfit patients with de novo AML and FLT3 mutations suggests that azacitidine can be combined successfully with FLT3 inhibitors with best overall response rates (ORR 60%¹², CR/CRi 67%¹³) seen when paired with gilteritinib. However, FLT3 tyrosine kinase inhibitors along with DAC10 and venetoclax produced even better CR/CRi outcomes (86%).¹³

Professor DiNardo and colleagues have investigated the combination of ivosidenib and venetoclax in 23 patients with IDH1 mutated newly diagnosed and secondary AML which demonstrated an ORR of 78.3% (95% CI, 56.3 – 92.5) with a CR rate of 60.9%  (95% CI, 38.5 – 80.3) and an OS at 12-month of 82% (95% CI, 58.8 – 92.8).¹⁴

In patients with the common TP53 mutated MDS/AML, promising CR rates (67% in MDS and 80% in AML) have been shown with the combination treatment of the p53 re-activator APR-246 plus azacitidine (n=20), with the most common ≥ grade three TEAEs being thrombocytopenia (n=5), neutropenia (n=6), and leukopenia (n=4).¹⁵

Professor DiNardo then discussed the use of the triplet combination treatment;azacitidine + venetoclax + pevonedistat, for newly diagnosed AML providing information about a trial that is currently ongoing¹⁶. Pevonedistat is a selective NEDD8 inhibitor that interferes with proteasome-mediated protein degradation and DNA repair, thereby causing apoptosis in malignant cells.¹⁷

Professor DiNardo finished with data demonstrating improved survival outcomes for patients with NPM1 mutated AML treated with an HMA plus venetoclax at three and five years (unpublished).

Courtney DiNardo concluded her presentation by highlighting the questions she felt should be considered when treating the older unfit patient with AML:

• SCT in older patients?
  • Is intensive chemotherapy required before performing SCT in older patients?
• What is the preferred regimen for patients with FLT3 and IDH mutant AML?
  • Are doublet combination sufficient (e.g. azacitidine + targeted treatment or azacitidine e + venetoclax) or should these patients receive a triplet combination?
• What mediates venetoclax relapse? Can we prevent them?
• Can we change the natural history of TP53 mutant AML?
• How do we best incorporate maintenance therapy, immunotherapy and/or monoclonal antibodies in the treatment of patients with AML?

The audience was then asked to reconsider the two case studies from the start of the presentation and how to treat them. There was a greater steer towards non-intensive induction chemotherapy for Patient one (78%), and for Patient two (56%).

The AML global portal interview with Courtney DiNardo can be found here and can be viewed below.