Allogeneic Hematopoietic Stem Cell Transplantation In Therapy-Related Myeloid Neoplasms (t-MN) of the Adult: Monocentric Observational Study and Review of the Literature.

BACKGROUND: Therapy related myeloid neoplasms (t-MN) occur due to direct mutational events of chemotherapeutic agents and radiotherapy. Disease latency, mutational events and prognosis vary with drugs categories.
METHODS: We describe a cohort of 30 patients, 18 females and 12 males, with median age of 52.5 years (range, 20 to 64), submitted to allogeneic stem cell transplantation (HSCT) in our department between September 1999 and March 2017. Patients had a history of solid tumour in 14 cases, haematological disease in 15 cases and both of them in one case. After a median of 36.5 months (range, 4 to 190) from first neoplasm, patients developed t-AML in 19 cases and t-MDS in 11 cases. Molecular abnormalities were detected in 5 patients, while karyotype aberrations were found in 17 patients. Patients received conventional chemotherapy in 14 cases, azacitidine in 10 cases and both of them in one case. Five patients were submitted to HSCT without previous treatment except for supportive therapy.
RESULTS: Seventeen patients obtained sustained CR after SCT, while 8 patients showed resistant or relapsed disease. The remaining five patients died early after SCT. At follow up time (May 2017) 13 patients were alive with a median OS of 48 months (range 3-195), while 17 patients died after a median of 4 months (range 1-27) by relapse mortality in 6 cases and non-relapse mortality in the other 11 patients.
CONCLUSIONS: Global OS was 43%. After SCT, 72.2% of patients with t-MN maintained a sustained CR.

Introduction

Therapy-related myeloid neoplasms are recognized as a separate entity in the World Health Organization (WHO) classification of haematological diseases.1 The incidence of therapy-related myeloid neoplasms (t-MN) continue to rise due to the relative prolongation of survival and cure related to chemo- and radio-therapy for primary malignancies, mostly breast cancer and lymphoproliferative diseases.2–7 The peak occurrence time of therapy-related acute myeloid leukemia/myelodysplastic syndrome is 3 to 5 years after prior cytotoxic treatment, while the risk decreases markedly after the first decade.8 At present, t-MN account for 10–20% of all malignant myeloid diseases.9 Factors associated with an increased risk of t-MN include exposure to alkylating agents, topoisomerase II inhibitors, radiation therapy,10–15 and older age at treatment, in addition to genetic susceptibility.16–21 t-MN after anthracyclines and/or topoisomerase II inhibitors are associated with occurrence of MLL translocation at 11q23 or RUNX1/AML1 at 21q22 after a median latency of 1 to 3 years without a prodromal phase. t-MN after alkylating agents have a median latency of 4 to 10 years and are often preceded by myelodysplasia. It is associated with unbalanced chromosome 5 and 7 abnormalities, complex karyotypes, and/or TP53 mutations. After radiation treatment, the highest risk for t-MN occurrence is registered at 2 years and appears to normalize after 10 to 15 years.22–24 Particularly, patients who received radiation to chest, pelvis and vertebrae for stomach, colorectal, liver, breast, endometrial, prostate, and kidney cancers seem to be at a significantly higher risk of developing t-MN.24,25 More recently, it came to light the potential role of various germline genetic factors in an individual’s susceptibility to t-MN, particularly for those variants that alter drug metabolism such as gene NQ01, glutathione-S-transferase,9,18,19,26 as well as those involved in DNA repair pathway such as BRCA, TP53 and MDM2.20,21

Clonal cytogenetic abnormalities are found in 75–90% of t-MN, and 46–70% of them are adverse karyotype including complex karyotype, deletion or loss of chromosome 5 and/or 7.3,4 Cytogenetics assessment is the principal prognostic factor for relapse rate and overall survival (OS).27–29 The heterogeneous treatments of therapy-related myeloid neoplasms, ranging from best supportive care to intensive chemotherapy, hypomethylating agents, and allogeneic stem cell transplantation, do not allow definite conclusions on the best treatment choice, particularly for elderly patients.30,31 Treatment of t-MN with conventional therapy is associated with a poor outcome in terms of survival (6 months),8,32 remission rate (28% to 50%) and duration of the remission.33–35 On the other hand, conventional chemotherapy might be a reasonable option for t-MN with favourable karyotype such as inv(16), t(16;16), t(15;17) or t(8;21), since the reported remission rate and the disease free survival are similar to those seen for the de novo counterpart.35,36 The introduction of new drugs such as azacitidine and decitabine has shown promising results in the management of t-MN with an acceptable toxicity profile also for frail patients, and with an overall response rate of approximately 40%.4,30,31,37,38–44

Allogeneic Stem Cell Transplantation for t-MN

Allogeneic haematopoietic stem cell transplantation (HSCT) represents the only potentially curative strategy, but it is not feasible for all patients due to age, comorbidities in elderly patients, poor organ reserve and high non-relapse mortality (NRM).4,45 The haematopoietic cell transplantation-specific comorbidity index (HCT-CI) was developed as a sensitive tool to measure the burden of comorbidities before HSCT and to predict both the risks of NRM and the probabilities of survival after HSCT.46 As reported by ElSawy et al.47 in the HCT-CI validation study, the three HCT-CI risk groups with score 0, 1–2, and ≥3 result in a NRM of 14%, 23%, and 39% with a survival of 74%, 61%, and 39%, respectively. Therefore, HSCT should be offer as a reliable option to fit patients with good performance status, intermediate and poor risk karyotype with suitable and available donor.9,27,28,48–50 With particularly interest to t-MN, the Center of International Bone Marrow Transplantation Research (CIBMTR) and the European Group for Bone and Marrow Transplantation (EBMT) extrapolated pre-transplant factors predicting post-HSCT outcome in these patients from larges study cohorts. CIBMTR conducted a large study cohorts on t-MN and proposed a prediction model of survival after allogeneic HSCT using the following four risk factors: age older than 35 years, poor-risk cytogenetics, t-AML not in remission or advanced t-MDS, donor other than an HLA-identical sibling or a partially or well-matched unrelated donor. Five-year survival for subjects with none, 1, 2, 3, or 4 of these risk factors was 50%, 26%, 21%, 10%, and 4%, respectively.27 Also the EBMT group28 reported that disease stage at transplant different from complete remission, abnormal cytogenetics (excluding t(8;21), inv(16) and t(15;17)) and patients’ age >40 years are the most significant factors predicting survival, relapse rate, disease-free survival (DFS) and NRM dividing patients into three risk groups: low, intermediate and high. Overall survival for the above-mentioned groups was 62%, 33% and 24%, respectively; DFS was 58% (low), 32% (intermediate) and 20% (high); NRM was 22% (low), 37% (intermediate) and 38% (high); finally, relapse rate was 20% (low), 31% (intermediate) and 32% (high) respectively.

We performed a review of the literature on therapy-related AML/MDS submitted to allogeneic stem cell transplantation excluding AML secondary to MDS progression. Detailed results concerning cohort size, median follow up, overall survival, NRM incidence, and relapse rate are depicted in Table 1. The reported outcomes for patients submitted to HSCT for therapy-related AML/MDS are very heterogeneous. Median OS ranges from 22% to 66%, with a NRM of 21 to 58% and a relapse rate of 26% to 42%.2,27,28,38,51–62

Table 1

Results of the review: outcomes of patients with therapy-related AML/MDS submitted to HSCT.

Author and year	N° of patients	Poor karyotype (%)	OS	NRM	Relapse rate	Median follow-up
Finke et al, 201651	79	53%	38%	23%	42%	7.5 ys
Tang et al, 201652	16	43%	66%	13%	20%	3.3 ys
Alam et al, 20152	65	50%	34%	31%	30–36%	5.9 ys
Liu et al, 201553	30	-	33%	-	34%	2 ys
Spina et al, 201254	29	59%	37%	32%	33%	3.7 ys
Zinke-Cerwenka et al, 201155	17	47%	47%	30%	24%	2.6 ys
Armand et al, 201056	24	50%	41%	17%	38%	2.8 ys
Litzow et al, 201027	868	26%	22%	48%	31%	5 ys
Kröger et al, 200928	461	42%	35%	37%	31%	1.8 ys
Nevill et al, 200857	24	46%	33%	30%	38%	4.5 ys
Chang et al, 200738	257	51%	33%	54%	33–36%	3.8 ys
Witherspoon et al, 200158	111	-	-	52–58%	26–40%	5 ys
de Witte et al, 200059	67	-	35%	46%	36%	5 ys
Yacoub-Agha et al, 200060	70	-	30%	49%	42%	7.9 ys
Anderson et al, 199761	46	25%	26%	44%	33%	5 ys
Ballen et al, 199762	18	50%	28%	50%	22%	3 ys
This report, 2017	30	32%	41%	44%	27%	2 ys

Monocentric Observational Study

Patients and disease characteristics

We retrospectively analyzed patients submitted to HSCT in our department and identified 30 patients with a diagnosis of therapy-related myeloid neoplasm (t-MN) transplanted between September 1999 and March 2017. Patients were 18 females (60%) and 12 males (40%) with a median age of 52.5 years (range, 20 to 64). Secondary neoplasm was acute myeloid leukemia (t-AML) in 19 cases (63%) and myelodysplasia (t-MDS) in 11 cases (37%). Data were collected through retrospective chart review and after institutional review board approval. The median time occurred from primary disease to t-MN occurrence was of 36.5 months (range, 4 to 190). Primary disease was hematologic in 15 cases (50%): Hodgkin’s disease (n=2), non-Hodgkin’s lymphoma (n=9), acute lymphoblastic leukemia (n=1), chronic lymphocytic leukemia (n=2) and acute myeloid leukemia (n=1). Fourteen patients (50%) had a previous diagnosis of solid tumor: medulloblastoma (n=1), breast (n=8), Ewing sarcoma (n=1), thyroid (n=1), bladder (n=2) and vagina/anus (n=1). One patient had a history of both haematological (non-Hodgkin’s lymphoma) and solid tumor (breast). Twelve patients (40%) had been previously treated with chemotherapy, 8 patients (26.7%) with chemotherapy and autologous transplantation, 2 (6.7%) patients with radiotherapy, one patient (3.3%) with radioiodine therapy and 7 patients (23.3%) with a combination of chemo- and radiotherapy. At t-MN diagnosis all patients had received a median of 2 lines of therapy (range, 1 to 6) for their primary malignancy. All patients were free of their primary malignancies at the time of transplantation.

Revised International Prognostic Scoring System (IPSS-R)63 was used to classify cytogenetics of t-MDS, while European Leukemia Net AML risk stratification by cytogenetics was used for AML.64 Karyotype was available for 28 out of 30 patients. Eleven patients (36.7%) had normal karyotype, three patients (10%) had a favourable karyotype, 5 patients (16.7%) had an intermediate-risk karyotype and 9 patients (30%) had an adverse-risk karyotype. Molecular cytogenetics analyses were available for 14 out of 30 patients: FLT3/ITD+ (n=2), CBFB/MYH11 (n=1), NPM1+ (n=1), NPM1 and FLT3/ITD double positivity (n=1), no abnormalities (n=9). A detailed description of primary neoplasms, treatment for primary neoplasm and t-HN is reported in Table 2. Transplant features and outcomes are depicted in Table 3.

Table 2

Detailed report of patients, primary and therapy-related disease and treatment.

Patients n.	Sex	Age	First neoplasia	Treatment for first neoplasm	Time to t-MN (months)	t-MN	Blasts count %	Molecular marker	Karyotype	Pre-HSCT treatment
1	Female	60	NHL	CHT+ASCT	18	t-MDS	20	N.A.	46,XX	Aza
2	Female	53	Breast cancer	CHT+RT	33	t-AML	20	N.A.	Hyperploid (93–94, XX), +G, −F, +C*	None
3	Female	49	Breast cancer	CHT	32	t-AML	23	Inv (16)	46, XX, Inv (16)	SD-CHT
4	Female	62	NHL	CHT+ASCT	120	t-MDS	N.A.	N.A.	46, XX, (−7)	Aza
5	Female	29	Breast cancer	CHT+RT	24	t-AML	N.A.	N.A.	46, XX	LD-CHT
6#	Male	33	AML	CHT+ASCT	72	t-MDS	N.A.	N.A.	45, XY, (−7)	Aza
7	Male	30	NHL	CHT+ASCT	60	t-MDS	15	N.A.	N.A.	None
8	Female	57	Breast cancer	RT	180	t-AML	58	NPM1+	46, XX	SD-CHT
9	Male	48	CLL	CHT	37	t-AML	20	None	46, XY, (−7)	SD-CHT
10	Female	36	Breast cancer	CHT+RT	48	t-AML	90	None	46, XX	Aza
11	Female	56	Breast cancer and NHL	CHT+RT	108	t-MDS	N.A.	N.A.	46, XX, (−7)	Aza
12	Female	48	Thyroid cancer	RIT	4	t-AML	20	NPM1+, FLT3+	N.A	SD-CHT
13	Male	48	Bludder cancer	CHT	48	t-AML	34	None	46, XY, t(3;3), (−7), (+8)	SD-CHT
14	Female	55	HL	CHT	36	t-AML	N.A.	N.A.	N.A.	SD-CHT
15	Male	55	NHL	CHT+ASCT	144	t-MDS	4	N.A.	46, XY	Aza
16	Female	57	Breast cancer	RT	24	t-AML	9	N.A.	46, XX, (+8)	None
17	Male	40	ALL	CHT	29	t-MDS	5	N.A.	45, XY (−7)	None
18	Female	53	Breast cancer	CHT+RT	120	t-AML	N.A.	N.A.	47, XY, (−11)(q14q23)	LD-CHT
19	Female	56	NHL	CHT+ASCT	24	t-MDS	6	None	N.A.	Aza
20	Female	41	Ewing sarcoma	CHT	17	t-AML	85	None	46, XX	SD-CHT
21	Male	62	NHL	CHT	61	t-MDS	5	N.A.	48, XXY (+8)	Aza
22	Female	39	Breast cancer	CHT+ASCT	30	t-AML	N.A.	N.A.	Hypoploid (42–44, XX)	SD-CHT
23	Female	55	NHL	CHT+RT	16	t-AML	91	None	46, XX, (−16), (+13)	SD-CHT
24	Male	53	NHL	CHT	39	t-AML	20	None	46, XY	Aza
25	Male	57	Bludder cancer	CHT	12	t-AML	30	FLT3+	N.A.	SD-CHT
26	Male	20	Medulloblastoma	CHT	190	t-MDS	8	None	46, XX, (−7p), (−1p), (−5q)	Aza
27	Male	59	NHL	CHT	12	t-AML	40	FLT3+	46, XY	SD-CHT
28	Male	64	CLL	CHT+ASCT	48	t-MDS	3	N.A.	46, XY (-20)	SD-CHT
29	Female	50	HL	CHT	17	t-AML	43	None	46, XX	SD-CHT+Aza
30	Female	52	Vagina-anus cancer	CHT+RT	100	t-AML	16	N.A.	N.A.	SD-CHT

Abbreviations: t-AML=therapy-related acute myeloid leukemia; t-MDS=therapy-related myelodysplastic syndrome; NHL=non-Hodgkin lymphoma; CLL=chronic lymphocytic leukemia; HL=Hodgkin lymphoma; ALL=acute lymphoblastic leukemia; CHT=chemotherapy; RT=radiotherapy; ASCT=autologous stem cell transplantation; Aza=azacitidine; SD-CHT=standard dose chemotherapy; LD-CHT=low dose chemotherapy; RIT: radioiodine therapy. N.A.=not available;

not otherwise specified deletion in the F group and duplication in the G and C group. # [Patients in question had an AML with t(8;21) as first neoplasia. Seven years after the last therapy (autologous stem cell transplantation), he developed a myelodysplasia with deletion of chromosome 7, while t(8;21) was not detected].

Table 3

Transplant for t-MN: features and outcomes.

Patient n.	Status at HSCT	HCT-CI	Time from t-HN to HSCT (months)	HSCT year	Donor	Stem cells source	Conditioning	GvHD prophylaxis	Disease response	DFS months	GvHD (acute or chronic)	Outcome	Cause of death	Survival months
1	refractory	9	13	2013	REL	PB	RIC	CSA+MTX	relapse	5	chronic	alive		48
2	untreated	4	2	2002	REL	PB	MAC	CSA+MTX	remission	173	chronic	alive		173
3	CR	3	6	2001	REL	PB	MAC	CSA+MTX	remission	195	both	alive		195
4	refractory	6	11	2014	REL	PB	RIC	CSA+MFA	refractory	0	acute	dead	NRM	3
5	refractory	3	3	1999	REL	PB	MAC	CSA+MTX	relapse	6	none	dead	RRD	6
6	PR	4	5	2013	MUD	PB	RIC	CSA+MFA+ATG	N.A.	1	acute	dead	NRM	1
7	untreated	4	4	2006	MUD	CB	RIC	CSA+MFA+ATG	remission	135	chronic	alive		135
8	CR	5	19	2015	REL	PB	MAC	CSA+MTX	remission	18	chronic	alive		18
9	refractory	7	4	2013	MUD	PB	RIC	CSA+MTX+ATG	refractory	0	none	dead	RRD	5
10	CR	4	9	2012	MUD	PB	MAC	CSA+MTX+ATG	remission	4	both	dead	NRM	4
11	CR	5	11	2012	MUD	PB	RIC	CSA+MFA	N.A.	1	none	dead	NRM	1
12	CR	3	19	2009	MUD	PB	MAC	CSA+MTX+ATG	N.A.	1	none	dead	NRM	1
13	refractory	3	3	2007	REL	PB	RIC	CSA+MFA	remission	6	acute	dead	NRM	6
14	CR	3	16	2009	REL	PB	MAC	CSA+MFA	remission	98	chronic	alive		98
15	CR	4	26	2011	MUD	PB	MAC	CSA+MTX+ATG	remission	8	both	dead	NRM	8
16	untreated	3	2	2016	REL	PB	MAC	CSA+MTX+ATG	remission	14	both	alive		14
17	untreated	5	4	2014	MUD	BM	MAC	CSA+MFA+ATG	remission	37	both	alive		37
18	refractory	3	7	2004	MUD	PB	MAC	CSA+MTX+ATG	relapse	12	none	dead	RRD	12
19	refractory	7	8	2016	HAPLO	BM	MAC	CSA+MFA+Cy	N.A.	1	none	dead	NRM	1
20	CR	3	7	2017	HAPLO	BM	MAC	CSA+MFA+Cy	remission	3	acute	alive		3
21	PR	4	12	2015	MUD	PB	MAC	CSA+MTX+ATG	remission	19	both	alive		19
22	refractory	3	9	2009	REL	PB	MAC	CSA+MTX	refractory	0	none	dead	RRD	4
23	CR	3	6	2009	REL	PB	MAC	CSA+MTX	remission	92	none	alive		92
24	CR	5	11	2013	MUD	PB	MAC	CSA+MTX+ATG	relapse	15	both	dead	RRD	16
25	refractory	3	6	2009	REL	PB	MAC	CSA+MTX	relapse	3	acute	dead	RRD	4
26	PR	6	14	2014	HAPLO	BM	MAC	CSA+MFA+Cy	remission	32	none	alive		32
27	CR	3	6	2011	REL	PB	RIC	CSA+MFA	remission	68	both	alive		68
28	untreated	3	36	2006	MUD	PB	RIC	CSA+Alemtuzumab	remission	27	chronic	dead	NRM	27
29	refractory	5	16	2011	MUD	PB	RIC	CSA+MTX+ATG	N.A.	1	acute	dead	NRM	1
30	CR	3	7	2011	MUD	PB	RIC	CSA+MFA	remission	4	acute	dead	NRM	4

CR=complete remission; PR=partial remission; REL=match related donor; MUD=match unrelated donor; Haplo=related haploidentical donor; PB=G-CSF-primed peripheral blood stem cells; BM=un-manipulated bone marrow stem cells; CB=un-manipulated cord blood stem cells; MAC=myeloablative conditioning; RIC=reduced intensity conditioning; CSA=cyclosporine A; MTX=methotrexate; MFA=mycophenolate mofetil; ATG=anti-lymphocytes globulin; Cy=post-transplant cyclophosphamide.

Statistical analysis

Overall survival and disease-free survival (DFS) were estimated using Kaplan-Meier product method, while for curves comparison log-rank test was applied. χ2 test and Fisher’s exact test were used to assess associations between categorical variables and OS, NRM, RRD, DFS. A competing risk analysis was performed to calculate the cumulative incidence of relapse-related death (RRD) and non-relapse mortality (NRM). For NRM, relapse was the competing event, and for relapse, NRM was the competing event. Fine and Gray’s method for cumulative incidence of RRD and NRM were used to compare different groups. Statistical analysis was realized using NCSS 10. A p-value ≤ 0.05 was considered statistically significant.

Results

Engraftment and GvHD

White blood cells count of ≥1.0×109/L and stable platelets count ≥20.0×109/L were reached at median day +21 (range, 11 to 130) and median day +15 (range, 10 to 45), respectively. Three patients died early before achieving stable engraftment.

Acute GvHD (aGvHD)65 occurred in 15 patients (50%) and global grading was as follows: grade I (n=3), grade II (n=5), grade III (n=6), and grade IV (n=1). Among them, three patients died because of aGvHD. Chronic GvHD (cGvHD)66 was diagnosed in 14 out of 23 patients surviving after day +100 (65%) and global scoring was as follows: mild (n=3), moderate (n=7) and severe (n=4). One of them died for cGvHD-related complications.

Response

Morphological bone marrow cytology was performed on day +30 after HSCT only in 25 patients because of early death in the others five. Three patients (12%) had a persistence of the underlying disease, whereas twenty-two patients achieved a CR (88%) on day +30. Among them, 5 patients (22.7%) experienced a relapse after a median time of 6 months (range, 3 to 15), while 17 patients (77.3%) maintained a CR after a median time of 27 months (range, 3 to 195). Median 2-ys DFS after HSCT was of 72.2% (95% CI 51.1 to 93.3) (Figure 1A).

Figure 1

Five-years outcomes of therapy-related AML/MDS after HSCT: 1A) Kaplan Meier for DFS; 1B) Kaplan Meier for OS; 1C) cumulative incidence of NRM; 1D) cumulative incidence of RRD; 1E) Kaplan Meier for OS according to cGvHD development; 1F) Kaplan Meier for OS according to relapse occurrence; 1G) cumulative incidence of NRM according to transplant conditioning regimen; 1H) cumulative incidence of RRD according to cGvHD development.

Overall survival, NRM and RRD

At the follow up data fixed on May 2017, 13 patients were alive after a median time of 48 months (range, 3 to 195), while 17 patients died after a median time of 4 months (range, 1 to 27). The causes of death were as follows: underlying disease (n=6), GvHD (n=3), EBV-related post-transplant lymphoproliferative disease (PTLD) (n=1) and infectious complications (n=7). The overall survival at 2 years after HSCT was of 40.5% (95% CI 22.1 to 58.9), whereas the cumulative incidence of NRM and RRD at 2 years was of 44.4% (95% CI 27.6 to 71.2) and 29.6% (95% CI 15 to 58.6), respectively (Figures 1B, 1C and 1D). No differences in terms of OS, NRM, RRD and DFS were seen stratifying patients according to underlying disease, disease status at transplant, previous treatment received, karyotype risk, patients and donor characteristics, stem cell source. An association was identified between OS and cGvHD development after HSCT, as well as between OS and relapse occurrence. Overall survival was higher in the group with cGvHD than those detected in the group without this complication (68% vs. 22%, p=0.018). Median OS was of 6 months (range, 4.6 to 7.4) in the group without cGvHD, while it was not reached in the group with cGvHD (p=0.0002, Figure 1E). An higher mortality was recorded in the group of patients who experienced a relapse of the underlying disease as compared with patients who did not relapsed after HSCT (67% vs. 13%, p=0.011). Median OS in the group relapsed after HSCT was of 5 months (range, 2.2 to 7.8) as compared to patients without relapse, for whom a median OS was not reached (p=0.004, Figure 1F). Relatively to NRM, an association was identified with the conditioning regimen: surprisingly, NRM was higher for patients who had received a reduced intensity conditioning as compared to those who had received a myeloablative one (p=0.046). Two-years cumulative incidence of NRM was of 74% (95% CI 49 to 100) after RIC transplant and 24% (95% CI 10 to 58) after ABL transplant (p=0.022, Figure 1G). Finally, also for RRD an association was found with cGvHD development after HSCT: among patients with cGvHD, a minor number of RRD was recorded as compared to patients who had not developed this complication (p=0.018). The cumulative incidence of RRD at 2 years after HSCT was of 9% (95% CI 1 to 59) for patients with cGvHD and 65% (95% CI 38 to 100) for patients without cGvHD (p=0.004, Figure 1H).

Two patients (6.7%) experienced a third tumor, in particular a breast cancer occurred thirteen years after HSCT and an EBV-related PTLD of the brain occurred eight months after HSCT.

Discussion

In the last two decades, many authors published results concerning different cohorts of patients with therapy-related acute myeloid leukemia or myelodysplasia submitted to allogeneic stem cell transplantation. An high heterogeneity in the percentage of OS (22% to 66%), NRM (21% to 58%) and relapse rate (26% to 42%) come to light from these experience.2,27,28,38,51–62 Each of these studies highlighted a different key point in this transplant setting, which might affect outcome after HSCT. The mainly predicting factor for OS resulted the karyotype and the recipient performance status at transplant.38,54,56 Patients achieving a CR before transplantation showed better outcomes, whereas multiple therapy lines increase organ damage as well as the incidence of neutropenia, infection events and the immunosuppression of the patient increase TRM.54,60,61 Patients at risk for treatment-related myeloid neoplasms should be followed closely and be considered for stem-cell transplantation early in the course of myelodysplasia.38,58,61 Considering the incremented risk of relapse according to blasts percentage, patients with secondary MDS should be direct to transplantation before the progression into AML, and if secondary AML occurs, they should be transplanted as soon as possible.61 For patients who did not achieved a CR pre-transplant, rapid transplantation, also considering alternative donor, could offer a reasonable outcome, reducing the risk of deterioration of the patient’s performance status. OS after HSCT in patients aged 60 years or above was very poor.50,51,67,68 Reduced intensity conditioning and conditioning with targeted busulphan dose38,51,58 might reduce TRM, especially for those patients with a reduced organ reserve. As reported for patients with de novo MDS,69 pre-transplant disease stage, cytogenetic risk group,57,56 type of therapy given for the original disease, transplant conditioning regimen, and patient age61 significantly affect relapse-free survival among patients with secondary MDS/t-AML.38 Concerning to stem cell source, peripheral blood instead of bone marrow appeared to reduce NRM38 and relapse rate38,57 and to improve OS.38 On the other hand, controversial data were reported relative to donor source impact on OS.2,27,38,53,70

In our cohort, global OS appeared to fit with those reported from several authors (40.5% vs 22–66%), whereas NRM appeared the major cause of death, even if the NRM rate was comparable to others data (44% vs 21–58%).2,27,28,38,51–62 Surprisingly, we observed an high DFS (72.2%) perhaps attributable to high cGvHD rate after HSCT, corresponding to an enhanced GvL effect. In fact, among patients with cGvHD a reduced RRD and an increased OS were registered. Graft-versus leukemia (GvL) effect, especially associated with chronic GvHD, improved DFS and OS also in adverse karyotype t-MN submitted to HSCT.71 Probably due to the small size of our study group, no differences in terms of post-transplant outcomes emerged dividing patients according to recipient age, previous treatment, disease status at transplant, karyotype, donor or stem cell source. Unexpectedly, we found a higher NRM among patients who had received a RIC transplant as compared to ABL, but no differences in performance status, pre-transplant risk score or disease status existed between the two groups.

An interesting feature revealed by our curves was that DFS reached a plateau approximately after the first year post HSCT, while OS reached its prolonged plateau after the second one. In fact, no relapse was ascertained after the first year post-HSCT, so that eighteen patients (56.7%) obtained and maintained a complete remission after HSCT. On the other hand, no deaths were recorded after the second year post-HSCT, with an OS of 40.5% at the follow up time.

Conclusions

The incidence of t-MN is increasing as more individuals survive treatment for a primary cancer diagnosis. At t-MN diagnosis,72 physicians should evaluate molecular and cytogenetic risk of the disease, performance status, age and comorbidities of patients, and should start HLA-typing to timely detect a suitable donor. Older patients with poor performance status should be offered clinical trials or best supportive care. For fit patients, molecular and cytogenetics stratification is crucial. t-APL might benefit from standard first line protocols. Favorable karyotype t-MN should be treated with standard induction chemotherapy followed by high dose cytarabine consolidation course. Normal karyotype t-MN could receive standard induction chemotherapy followed by HSCT while poor molecular karyotype t-MN should be encouraged to participate in prospective clinical trials specifically designed and they should be considered early for allogeneic HCT.51 Upfront HSCT could be offered to patients with low blast count and poor performance status.