Transfusion-free interval is associated with improved survival in patients with higher-risk myelodysplastic syndromes engaged in routine care

Figures & data

Table 1. IPSS-R classification for MDS [4].

Parameter	Score
	0	0.5	1.0	1.5	2.0	3	4
Cytogenetics^a	Very good	–	Good	–	Intermediate	Poor	Very poor
Marrow blasts (%)	≤2	–	>2 to <5	–	5–10	>10	–
Hgb (g/dL)	≥10	–	8 to <10	<8	–	–	–
Platelets (×10^9/L)	≥100	50 to <100	<50	–	–	–	–
ANC (×10^9/L)	≥0.8	<0.8	–	–	–	–	–
Sum of scores above for individual parameters = risk category
Score				Risk category
0–1.5				Very low
2–3				Low
3.5–4.5			TARGET POPULATION	Intermediate
5–6			FOR ANALYSIS	High
>6				Very high

a Very good (−Y, del[11q]); good (normal, del[5q], del[12p], del[20q], double including del[5q]); intermediate (del[7q], +8, +19, i[17q], any other single or double independent clones); poor (−7, inv[3]/t[3q]/del[3q], double including −7/del[7q], or complex [i.e. 3 abnormalities]); very poor (>3 abnormalities).

ANC: absolute neutrophil count; Hgb: hemoglobin.

Figure 1. Study design. ^aMDS chemotherapy includes the following: azacitidine, decitabine, lenalidomide, ATG, cyclosporine, cytarabine, idarubicin, daunorubicin, fludarabine, clofarabine and topotecan, or alloSCT. ^bMDS-specific supportive care includes the following: erythrocyte/platelet transfusions; thrombopoietic-, erythropoietic-, and granulocyte-stimulating agents, and hydroxyurea. AML: acute myeloid leukemia; alloSCT: allogeneic stem cell transplant; ATG: anti-thymocyte globulin; HR-MDS: higher-risk myelodysplastic syndrome; ICD-9/ICD-10: International Classification of Diseases 9th/10th Revision; IDN: integrated delivery network; MDS: myelodysplastic syndrome; SCT: stem cell transplant.

Table 2. Baseline patient characteristics and treatment characteristics.

Type of therapy^a	Overall N = 218	Azacitidine n = 138	Decitabine n = 54	Induction-type n = 9	Lenalidomide n = 17
Baseline characteristics
Age, n (%)
≥65 years	177 (81.2)	108 (78.3)	46 (85.2)	6 (66.7)	17 (100.0)
Gender, n (%)
Male	145 (66.5)	94 (68.1)	36 (66.7)	5 (55.6)	10 (58.8)
Race, n (%)
African American	6 (2.8)	1 (0.7)	3 (5.6)	1 (11.1)	1 (5.9)
Asian	1 (0.5)	0 (0)	1 (1.9)	0 (0)	0 (0)
Caucasian	206 (94.5)	135 (97.8)	48 (88.9)	7 (77.8)	16 (94.1)
Other/unknown	5 (2.3)	2 (1.5)	2 (3.7)	1 (11.1)	0 (0)
Ethnicity, n (%)
Hispanic	4 (1.8)	3 (2.2)	1 (1.9)	0 (0)	0 (0)
Not Hispanic	192 (88.1)	128 (92.8)	43 (79.6)	6 (66.7)	15 (88.2)
Unknown	22 (10.1)	7 (5.1)	10 (18.5)	3 (33.3)	2 (11.8)
Region, n (%)
Midwest	114 (52.3)	70 (50.7)	32 (59.3)	3 (33.3)	9 (52.9)
Northeast	20 (9.2)	12 (8.7)	6 (11.1)	1 (11.1)	1 (5.9)
South	7 (3.2)	5 (3.6)	1 (1.9)	0 (0)	1 (5.9)
West	63 (28.9)	40 (29.0)	15 (27.8)	3 (33.3)	5 (29.4)
Other/unknown	14 (6.4)	11 (8.0)	0 (0)	2 (22.2)	1 (5.9)
Payer type, n (%)
Commercial only	49 (22.5)	34 (24.6)	11 (20.4)	3 (33.3)	1 (5.9)
Medicare only	100 (45.9)	65 (47.1)	21 (38.9)	2 (22.2)	12 (70.6)
Medicaid only	4 (1.8)	4 (2.9)	0 (0)	0 (0)	0 (0)
Mixed payer	44 (20.2)	28 (20.3)	11 (20.4)	2 (22.2)	3 (17.6)
Other payer	2 (0.9)	0 (0)	1 (1.9)	1 (11.1)	0 (0)
Uninsured	2 (0.9)	1 (0.7)	1 (1.9)	0 (0)	0 (0)
Unknown	17 (7.8)	6 (4.4)	9 (16.7)	1 (11.1)	1 (5.9)
Diagnosis period, n (%)
2008–2011	59 (27.1)	26 (18.8)	23 (42.6)	4 (44.4)	6 (35.3)
2012–2015	159 (72.9)	112 (81.2)	31 (57.4)	5 (55.6)	11 (64.7)
Baseline CCI score, n (%)
0	96 (44.0)	57 (41.3)	27 (50.0)	6 (66.7)	6 (35.3)
1	46 (21.1)	30 (21.7)	11 (20.4)	2 (22.2)	3 (17.7)
≥2	76 (34.9)	51 (37.0)	16 (29.6)	1 (11.1)	8 (47.1)
Presence of select Charlson comorbidities at baseline, n (%)
Myocardial infarction	17 (7.8)	12 (8.7)	2 (3.7)	2 (22.2)	1 (5.9)
Congestive heart failure	29 (13.3)	21 (15.2)	6 (11.1)	1 (11.1)	1 (5.9)
Chronic pulmonary disease	50 (22.9)	33 (23.9)	12 (22.2)	1 (11.1)	4 (23.5)
Peripheral vascular disease	33 (15.1)	20 (14.5)	11 (20.4)	0 (0)	2 (11.8)
Mild liver disease	9 (4.1)	8 (5.8)	1 (1.9)	0 (0)	0 (0)
Connective tissue disease	5 (2.3)	3 (2.2)	1 (1.9)	0 (0)	1 (5.9)
Diabetes	49 (22.5)	31 (22.5)	10 (18.5)	1 (11.1)	7 (41.2)
Diabetes with end-organ damage	21 (9.6)	13 (9.4)	5 (9.3)	1 (11.1)	2 (11.8)
Moderate to severe renal disease	28 (12.8)	17 (12.3)	8 (14.8)	0 (0)	3 (17.7)
Cerebrovascular disease	24 (11.0)	14 (10.1)	9 (16.7)	0 (0)	1 (5.9)
Bone marrow blast count^b, n (%)
Unknown	143 (65.6)	88 (63.8)	38 (70.4)	6 (66.7)	11 (64.7)
≤2%	9 (4.1)	4 (2.9)	5 (9.3)	0 (0)	0 (0)
2% to <5%	14 (6.4)	11 (8.0)	2 (3.7)	0 (0)	1 (5.9)
5% to 10%	30 (13.8)	20 (14.5)	5 (9.3)	0 (0)	5 (29.4)
>10%	22 (10.1)	15 (10.9)	4 (7.4)	3 (33.3)	0 (0)
Hemoglobin^b, n (%)
Unknown	14 (6.4)	7 (5.1)	4 (7.4)	2 (22.2)	1 (5.9)
≥10 gm/dL	38 (17.4)	23 (16.7)	11 (20.4)	2 (22.2)	2 (11.8)
8 to <10 gm/dL	101 (46.3)	63 (45.7)	25 (46.3)	3 (33.3)	10 (58.8)
<8 gm/dL	65 (29.8)	45 (32.6)	14 (25.9)	2 (22.2)	4 (23.5)
Platelets^b, n (%)
Unknown	6 (2.8)	2 (1.4)	4 (7.4)	0 (0)	0 (0)
≥100 K/L	66 (30.3)	44 (31.9)	12 (22.2)	3 (33.3)	7 (41.2)
50 to <100 K/L	53 (24.3)	28 (20.3)	17 (31.5)	1 (11.1)	7 (41.2)
<50 K/L	93 (42.7)	64 (46.4)	21 (38.9)	5 (55.6)	3 (17.7)
Neutrophils^b, n (%)
Unknown	17 (7.8)	9 (6.5)	5 (9.3)	1 (11.1)	2 (11.8)
≥0.8 K/L	134 (61.5)	88 (63.8)	29 (53.7)	5 (55.6)	12 (70.6)
<0.8 K/L	67 (30.7)	41 (29.7)	20 (37.0)	3 (33.3)	3 (17.7)
Received transfusions at baseline^c, n (%)	57 (26.2)	42 (30.4)	10 (18.5)	2 (22.2)	3 (17.7)
Received any supportive care at baseline^d, n (%)	62 (28.4)	44 (31.9)	12 (22.2)	3 (33.3)	3 (17.7)
Follow-up treatment characteristics
Received supportive care during 1LT^e, n (%)
Any	180 (82.6)	119 (86.2)	43 (79.6)	6 (66.7)	12 (70.6)
CSFs	54 (24.8)	32 (23.2)	20 (37.0)	2 (22.2)	0 (0)
ESAs	46 (21.1)	25 (18.1)	13 (24.1)	1 (11.1)	7 (41.2)
Transfusions	157 (72.0)	109 (79.0)	36 (66.7)	4 (44.4)	8 (47.1)
Time (days) from diagnosis to initiation of 1LT, median (IQR)	21 (10, 66)	21 (11, 52)	20 (11, 60)	13 (3, 88)	41 (3, 189)
Duration of 1LT, median in days (IQR)	101 (36, 211)	117 (58, 205)	102 (33, 263)	29 (29, 35)	33 (29, 79)
Follow-up time, median in months (IQR)	9.4 (4.2, 17.4)	8.6 (4.4, 16.5)	12.8 (4.2, 22.1)	3.6 (1.5, 8.7)	13.3 (9.8, 20.4)

a Ten patients receiving azacitidine, three patients receiving decitabine, and one patient receiving induction-type chemotherapy also received SCT at any time during follow-up.

b Baseline levels were relative to the 12 months prior to and up to 30 days post-index diagnosis date, with the exception of hemoglobin, which was only prior to index diagnosis date; the value closest to the index diagnosis date was used.

c This is relative to platelet or red blood cell transfusions recorded in the 12 months prior to index diagnosis date.

d This includes MDS-directed supportive care (i.e. erythrocyte and platelet transfusions, erythrocyte- and platelet-stimulating agents, CSFs, and hydroxyurea) and is relative to the 12 months prior to index diagnosis date.

e This includes selected supportive care drugs/procedures of interest that occurred during 1LT or either 60 days prior to initiation of 1LT or 60 days after the last treatment of 1LT.

1LT: first-line therapy; CCI: Charlson comorbidity index; CSF: colony-stimulating factor; dL: deciliter; ESA: erythropoietin-stimulating agent; gm: gram; IQR: interquartile range; K: thousand; L: liter; MDS: myelodysplastic syndrome.

Figure 2. Prescribing patterns for patients receiving second-line therapy. ^aInduction-type treatment: an antimetabolite (i.e. cytarabine, fludarabine) ± an anthracycline (i.e. idarubicin or daunorubicin). ^bCytotoxic chemotherapy: any non-HMA chemotherapy (i.e. fludarabine ± another agent, clofarabine, cytarabine ± daunorubicin). ^cOf the lenalidomide patients, n = 4 were lenalidomide alone and n = 2 were in combination with azacitidine. HMA: hypomethylating agent.

Figure 3. Progression-free survival (A) and overall survival (B) in patients with and without a TFI receiving first-line therapy (p < .0001 for both). TFI: transfusion-free interval.

Table 3. Multivariate analysis of 2-year PFS^a,b and OS^a,b among HR-MDS patients with and without a TFI.

Parameters	HR (95% CI)	p Value
PFS at 2 years
TFI^c vs. no TFI	0.27 (0.16, 0.44)	<.0001
Age ≥65 vs. age <65	2.21 (1.01, 4.83)	.0475
CCI =1 vs. CCI =0	0.48 (0.24, 0.99)	.0471
CCI ≥2 vs. CCI =0	1.42 (0.75, 2.67)	.2774
Baseline supportive care used^d vs. not used	1.47 (0.84, 2.56)	.1741
Baseline hemoglobin ≥8 g/dL vs. <8 g/dL	1.79 (0.08, 2.99)	.0251
OS at 2 years
TFI^c vs. no TFI	0.34 (0.19, 0.60)	.0002
Age ≥65 vs. age <65	2.97 (1.11, 7.94)	.0300
CCI =1 vs. CCI =0	0.77 (0.33, 1.85)	.5641
CCI ≥2 vs. CCI =0	2.31 (1.13, 4.74)	.0218
Baseline supportive care use^d vs. not	2.45 (1.26, 4.76)	.0080

a Patients receiving SCT alone were excluded from this analysis; observations were censored at loss to follow-up or end of study period.

b Adjusted for TFI, age (<65 years vs. ≥65 years), gender, race, region, payer type, CCI (0, 1, ≥2), time to first-line therapy (in months), year of diagnosis (continuous), any baseline supportive care, baseline absolute neutrophil count, baseline hemoglobin, baseline platelets.

c TFI was defined as having a ≥60-day timeframe post-initiation of 1LT with no platelet or erythrocyte transfusions.

d MDS-directed supportive care use in baseline was composed of erythrocyte/platelet transfusions; thrombopoietic-, erythropoietic-, and granulocyte-stimulating agents; and hydroxyurea in the 12 months prior to index diagnosis date.

CCI: Charlson comorbidity index; CI: confidence interval; HR: hazard ratio; MDS: myelodysplastic syndromes; OS: overall survival; PFS: progression-free survival; SCT: stem cell therapy; TFI: transfusion-free interval.

Greenberg PL, Tuechler H, Schanz J, et al. Revised International Prognostic Scoring System for myelodysplastic syndromes. Blood. 2012;120:2454–2465.

 PubMed Web of Science ®Google Scholar