Unrelated Donor Hematopoietic Cell Transplantation for Non-Hodgkin Lymphoma: Long-term Outcomes

Abstract

We analyzed the outcomes of 283 patients receiving unrelated donor allogeneic hematopoietic cell transplantation for non-Hodgkin lymphoma (NHL) facilitated by the Center for International Blood & Marrow Transplant Research /National Marrow Donor Program (CIBMTR/NMDP) between 1991 and 2004. All patients received myeloablative conditioning regimens. The median follow-up of survivors is 5 years. Seventy-three (26%) patients are alive. The day 100 probability of death from all causes is estimated at 39%. The cumulative incidence of developing grade III-IV acute graft-versus-host disease (GVHD) at day 100 is 25%. The estimated five-year survival and failure free survival are 24% (95% CI; 19–30) and 22% (95% CI; 17–28) respectively. Factors adversely associated with overall survival included increasing age, decreased performance status, and refractory disease. Follicular lymphoma and Peripheral T-cell lymphoma had improved survival compared to aggressive B-cell lymphomas. Factors adversely associated with progression free-survival included performance status, histology and disease status at transplant. Long-term failure-free survival is possible following unrelated donor transplantation for NHL, although early mortality was high in this large cohort.

INTRODUCTION

In 1993 the NMDP reported on 462 patients undergoing unrelated donor hematopoietic stem cell transplantation (HCT). The large majority of these patients suffered from leukemia, only 8 patients with lymphoma were included (1). Since then, unrelated donor HCT has become a standard form of treatment for patients with acute leukemia (2–4) . Advances in HLA-typing have permitted more accurate identification of compatible donor-recipient pairs and many recent series suggest that the outcomes of HLA-identical unrelated donor transplantation are virtually identical to that of HLA-identical sibling transplantation for leukemia (5–7). At the same time, allogeneic sibling HCT has become more widely used in lymphoma. It is widely accepted as an excellent treatment for relapsed follicular lymphoma (8–11), and is often used as an alternative to autologous transplant in large cell lymphoma (12–15), mantle cell lymphoma (16, 17), T-cell lymphoma (18–21) and in high grade lymphomas (22, 23), and in patients with lymphoma who have failed prior autologous transplantation (24). Unrelated donor HCT is increasingly used in NHL (8, 12, 16, 25–28), and single institution studies report similar outcomes after related and unrelated donor transplantation for lymphoma (29), but large series are lacking. This report describes cure rates and treatment complications in a large patient cohort with very high risk characteristics undergoing unrelated allogeneic HCT for NHL facilitated by the NMDP. Since the large majority of the patients received TBI based conditioning, this dataset does not address questions regarding relative superiority of various transplant conditioning regimens.

PATIENTS AND METHODS

Data Sources

The CIBMTR is a research affiliation of the International Bone Marrow Transplant Registry (IBMTR), Autologous Blood and Marrow Transplant Registry (ABMTR) and the National Marrow Donor Program (NMDP) that comprises a voluntary working group of more than 450 transplant centers worldwide that contribute detailed data on consecutive allogeneic and autologous transplants to a Statistical Center at the Health Policy Institute of the Medical College of Wisconsin in Milwaukee or the NMDP Coordinating Center in Minneapolis. Participating centers are required to report all consecutive transplants; compliance is monitored by on-site audits. Subjects are followed longitudinally, with yearly follow-up. Computerized checks for errors, physicians’ review of submitted data and on-site audits of participating centers ensure data quality. Observational studies conducted by the CIBMTR are done with a waiver of informed consent and in compliance with HIPAA regulations as determined by the Institutional Review Board and the Privacy Officer of the Medical College of Wisconsin.

Patients

The outcomes of patients with NHL who underwent myeloablative unrelated allogeneic bone marrow or peripheral blood HCT facilitated by the NMDP between 1991 and 2004 are reported. The policies and procedures of the NMDP have been described previously (2, 3). All donors signed written statements of informed consent prior to donation. Patients with a prior autologous hematopoietic stem cell transplant and recipients of cord blood grafts (n=5) were excluded from analysis. Central Histology Review was not performed.

Study Endpoints

Outcomes analyzed included engraftment, acute and chronic graft versus host disease (GVHD), transplant-related mortality (TRM), relapse/progression, progression-free survival (PFS) and overall survival (OS). The incidence and stage of acute skin, liver, and intestinal GVHD were measured by standard criteria (30). Chronic GVHD was classified according to the standard criteria in use prior to the recent consensus statement (31). Lymphoma histology was classified according to the WHO classification (32). The day of engraftment was defined as the first of three consecutive days on which the absolute neutrophil count (ANC) exceeded 0.5 × 10⁹/L or time to neutrophil count > 3 × 10⁹/L on one occasion.

TRM was defined as death within 28 days post-transplant or death without lymphoma-progression, and summarized using the cumulative incidence estimate with progression or relapse as the competing risk. Progression was defined as progressive lymphoma post transplant or lymphoma-recurrence. It could follow a period of “stable” disease post transplant, or a partial or complete remission. Progression represents new or larger areas of lymphoma (≥25% increase in largest diameter) compared to the state of best post transplant response. Progression was summarized by the cumulative incidence estimate with TRM as the competing risk. For PFS, subjects were considered treatment-failures at the time of lymphoma-progression or death from any cause. Subjects alive without evidence of lymphoma-progression were censored at last follow-up and the PFS event was summarized by a survival curve. The OS interval variable was defined as time from the date of transplant to the date of death or last contact and summarized by a survival curve.

A variety of HLA typing methods were utilized over the period of study for matching donor recipient pairs. These included serologic typing, or molecular typing techniques. Donor recipient pairs were therefore reclassified as well matched, partially matched or mismatched according to recent criteria proposed by Weisdorf et al (33)

Statistical Analysis

Subject-, disease-, and transplant-related variables for subjects were described. Univariate probabilities of developing TRM and lymphoma-relapse/progression were calculated using cumulative incidence curves to accommodate corresponding competing risks (33). Probabilities of overall and progression-free survival were calculated using Kaplan-Meier estimator (34). Confidence intervals (CI) were calculated with a log-transformation.

The effect of pre-transplant variables on the outcomes of TRM, lymphoma-progression, PFS and survival after transplantation, was compared using Cox proportional hazards model incorporating pre-transplant variables of interest (35). Covariates which influenced outcomes were identified by stepwise forward selection multivariate model. Any covariate with a p-value ≤0.05 was considered significant. The proportionality assumption for Cox-regression was tested by adding a time-dependent covariate for each risk factor and each outcome. Tests indicated that all variables met the proportional-hazards assumption. Results were expressed as relative risks (RR) or the relative rate of occurrence of the event. The following variables (summarized in Table 3) were considered in multivariate analyses: age at transplant, sex, Karnofsky Performance Status (KPS) at transplant, lymphoma histology, disease status and chemosensitivity at transplant, time from diagnosis to transplant, TBI based conditioning, T cell depletion, donor – recipient HLA match, donor-recipient CMV status and year of transplant. There were no statistically significant center effects (36). Analyses were performed using SAS software, version 8.2 (SAS Institute).

Table 3.

Variables tested in Cox proportional hazards regression models.

Patient-related variables:

Age: ≤20* vs 21–40 vs ≥40

Sex: male* vs female

Karnofsky performance status: <90%* vs ≥90%

Disease-related variables

Histology: Follicular* vs DLCL/Immunoblastic vs Lymphoblastic/Burkitts/Burkitt-like vs

Mantle Cell vs Peripheral T-cell vs others

Disease status at transplant: Chemosensitive CR* vs chemosensitive PR vs chemoresistant

Treatment-related variables:

Time from diagnosis to transplant: TBD (will look for optimal cutoff point)

TBI conditioning: no* vs yes

T-cell depletion: no* vs yes

Type of donor: well-matched* vs others

Donor age: 11–20* vs 21–30 vs 31–40 vs 41–50 vs 51–60

Donor type: well matched vs others

CMV status: +/+ vs +/− vs −/+ vs −/− vs not tested

Year of transplant: 1991–1994* vs 1995–1999 vs 2000–2004

^*Reference group

RESULTS

Patient Characteristics

Two hundred-eighty three NHL patients were identified who received unrelated allogeneic bone marrow transplants at 63 transplant centers between 1991 and 2004. Fifty six percent of these transplants occurred between 2000 and 2004. The median follow-up for surviving patients is 5.1 years (0.7–12.1). The median time interval between diagnosis and transplant was 1.4 (range 0.9 to 16) years. Patient characteristics are summarized in Table 1.

Table 1.

Characteristics of patients who underwent matched unrelated donor allogeneic hematopoietic stem cell transplant with myeloablative conditioning for NHL and reported to the CIBMTR between 1991–2004.

Variable	N eval	N (%)
Number of patients		283
Number of centers		63
Age, median (range), years	283	37 (2 – 65)
Age at transplant, years	283
≤10		16 ( 6)
11–20		34 (12)
21–30		62 (22)
31–40		58 (20)
41–50		84 (30)
51–60		27 ( 9)
≥61		2 ( 1)
Male sex	283	188 (66)
Karnofsky score pre-transplant	270
<90		100 (37)
≥90		170 (63)
Histology	283
Follicular		62 (22)
DLCL/Immunoblastic		73 (26)
Lymphoblastic/Burkitts/Burkitt-like		68 (24)
Mantle Cell		20 ( 7)
Peripheral T-cell		19 ( 7)
Othersa		41 (14)
Disease status at transplant	269
CR1		41 (15)
CR2+		30 (11)
PIF resistant		74 (28)
PIF sensitive		57 (21)
REL resistant		27 (10)
REL sensitive		40 (15)
Extranodal sites of disease at transplant	51
1		5 (10)
2		1 ( 2)
No extranodal involvement		45 (88)
Disease stage at diagnosis	283
I-II		59 (21)
III-IV		206 (73)
Unknown		18 ( 6)
Interval from diagnosis to transplant, median (range), months	283	17 (1 – 192)
Conditioning regimen	283
TBI + Cy ± other		212 (75)
TBI ± other		26 ( 9)
Bu + Cy ± other		38 (13)
Cy ± other		3 ( 1)
Othersb		4 ( 2)
Fludarabine	106
Yes		5 ( 5)
No		101 (95)
T-cell depletionc	283
Yes		78 (28)
No		205 (72)
Donor age, median (range), years	282	36 (20 – 54)
Donor age at transplant, years	282
11–20		7 ( 2)
21–30		76 (27)
31–40		121 (43)
41–50		70 (25)
51–60		8 ( 3)
Type of donord	283
Well-matched		153 (54)
Partially matched		111 (39)
Mismatched		17 ( 6)
Unknown		2 ( 1)
Graft type	283
Bone Marrow		223 (79)
Peripheral Blood		60 (21)
Donor-recipient sex match	283
M-M		132 (47)
M-F		57 (20)
F-M		56 (20)
F-F		38 (13)
Donor/Recipient CMV status	283
+/+		31 (11)
+/−		24 ( 8)
−/+		82 (29)
−/−		109 (39)
Not tested		37 (13)
GVHD prophylaxis	283
MTX + CsA ± other		127 (45)
CsA ± other		12 ( 4)
MTX + FK506 ± other		76 (27)
FK506 ± other		12 ( 4)
T-cell depletion ± other		52 (18)
Othersd		4 ( 2)
Year of transplant	283
1991–1994		11 ( 4)
1995–1999		114 (40)
2000–2004		158 (56)
Median follow-up of survivors, months	73	61 (8 – 145)

^aOthers histology includes (N=41):

1. NHL, NOS (n=3)

2. NHL other, not specified (n=35)

3. Extranodal marginal zone B-cell of MALT (n=1)

4. Nodal marginal zone B-cell (n=2)

^bOther conditioning regimen includes (N=4):

1. Bu + L-PAM (n=2)

2. Carboplatin + Thiotepa + other (n=2)

^cT-cell depletion includes:

1. 26 received in vivo T-cell depletion with ATG or Campath.

2. 40 grafts were in vitro T-cell depleted with various antibody based methods.

3. 12 grafts were in vitro T-cell depleted with antibody based methods combined with in-vivo administration of ATG.

^dType of donor defined as:

Well-matched includes: 1,7,10,16 (see below)

Partially matched includes: 2,3,8,11,13,17,21 (see below)

Mismatched includes 4,5,6,9,12,14,18,22,23,24 (see below)

	HLA Groups	N
1	Matched 8/8 at high-res HLA-A, -B, -C and –DRB1	112
2	Single allele MM (7/8) at high-res HLA-A, -B, -C and -DRB1	17
3	Single antigen MM (7/8) at high-res HLA-A, -B, -C and -DRB1	26
5	≥2 MM with 1 antigen MM (<7/8) at high-res HLA-A, -B, -C and -DRB1	13
6	≥2 MM with 2 or more antigen MM (<7/8) at high-res HLA-A, -B, -C and -DRB1	2
7	Matched 8/8 at high-res HLA-A, -B and -DRB1 and low-res at HLA-C	5
8	Single MM (7/8) at high-res HLA-A, -B and -DRB1 and low-res at HLA-C	2
9	≥2 MM (<7/8) at high-res HLA-A, -B and -DRB1 and low-res at HLA-C	2
10	Matched 8/8 at low-res HLA-A, -B and -C and high-res at HLA-DRB1	30
11	Single MM (7/8) at low-res HLA-A, -B and -C and high-res at HLA-DRB1	7
16	Matched 6/6 at high-res HLA-A, -B and -DRB1 (HLA-C unknown)	6
21	Matched 6/6 at low-res HLA-A and -B and high-res at HLA-DRB1 (HLA-C unknown)	59

^d

Other GVHD prophylaxis includes (N=4)

1. MTX ± other (n=1)

2. GVHD given, not specified (n=2)

3. None (n=1)

Follow-up completeness index = 98%

The median age at the time of transplant was 37 (range 2–65) years. Sixty-six percent of patients were male. Twenty-two percent of patients had follicular lymphoma, 26% diffuse large B cell lymphoma, 24% lymphoblastic, Burkitt’s or Burkitt’s like lymphoma, 7% mantle cell lymphoma, and 7% had peripheral T-cell lymphoma. Fourteen percent were classified as other. This group includes patients where no specific histology was provided, but also some with rare histologies (Table 1. footnote a). Thirty patients were reported as having transformed lymphoma. For analysis they were categorized under the histological diagnosis at the time of transformation. Only 15% of patients underwent transplant in first complete remission and 11% in second or subsequent remission. The majority of patients were not in remission at transplant – 49% with primary induction failure, 15% with sensitive relapse and 10% with resistant relapse. Overall, 38% patients were chemotherapy refractory at the time of transplant. The large majority of patients (84%) received TBI containing conditioning regimens. The donor grafts were non-T-depleted in 72% of cases and they were mostly bone marrow grafts. Standard GVHD prophylaxis i.e. a calcineurin inhibitor plus methotrexate was used in 72% of cases. Donors were well matched in 54%, partially matched in 39% and mismatched in 6% of cases. The matching for 2 donors (1%), was unknown.

Outcomes

Outcomes are summarized in Table 2. Ninety percent of patients engrafted and almost all of these had reached an ANC of >0.5 by day 28. The overall 100 day mortality was 39%. Treatment related mortality was 44% by one year and increased to 52% by five years. The cumulative incidence of acute grade 3–4 GVHD was 25% and the cumulative incidence of chronic GVHD increased to 32% at 5 years. The risk for disease progression was 25% at 1 year with no change in subsequent years. Progression free survival at one year was 31% and decreased to 22% at 5 years. Overall survival was 36% at one year and decreased to 24% at 5 years. (Figure 1)

Table 2.

Univariate probabilities of transplant outcomes among patients who underwent matched unrelated donor allogeneic hematopoietic stem cell transplant with myeloablative conditioning for NHL.

Outcome event	N	Prob (95% CI)a
100 day mortality	283	39 (33 – 45)
ANC>0.5 × 109/L	283
@ 28 days		89 (86 – 93)
@ 100 days		90 (87 – 93)
Acute GVHD @ 100 days, grades (3–4)	281	25 (20 – 30)
Chronic GVHD	280
@ 1 year		27 (22 – 32)
@ 3 years		31 (26 – 37)
@ 5 years		32 (26 – 37)
TRM	283
@ 1 year		44 (38 – 49)
@ 3 years		49 (43 – 55)
@ 5 years		52 (45 – 57)
Progression/Relapse	283
@ 1 year		25 (20 – 30)
@ 3 years		26 (21 – 32)
@ 5 years		26 (21 – 32)
PFS	283
@ 1 year		31 (26 – 37)
@ 3 years		25 (20 – 30)
@ 5 years		22 (17 – 28)
Overall survival	283
@ 1 year		36 (31 – 42)
@ 3 years		28 (22 – 33)
@ 5 years		24 (19 – 30)

^aProbabilities of neutrophil, acute and chronic GVHD, TRM, and progression/relapse were calculated using the cumulative incidence estimate. 100-day mortality, PFS and overall survival were calculated using the Kaplan-Meier product limit estimate.

Figure 1.

Overall Survival and Progression Free Survival after unrelated donor allogeneic hematopoietic stem cell transplant for NHL

Multivariate analysis

Multivariate analyses for the outcomes of TRM, Relapse/Progression, PFS and OS are summarized in Table 4–Table 7.

Table 4.

Multivariate analysis comparing treatment-related mortality among patients who underwent matched unrelated donor allogeneic hematopoietic stem cell transplant with myeloablative conditioning for NHL.

Variables:	N	Relative Risk (95% CI)	P-value
Age at transplant
≤20	46	1.00a	Pboverall = 0.003
21–40	105	1.76 (0.97 – 3.20)	0.062
≥40	105	2.82 (1.51 – 5.28)	0.001
Karnofsky performance status
≥90	162	1.00a
<90	94	1.53 (1.07 – 2.18)	0.018
Histology
Follicular	52	1.00a	Pboverall = 0.021
DLCL/Immunoblastic	70	1.91 (1.15 – 3.16)	0.012
Lymphoblastic/Burkitts/Burkitt-like	61	1.97 (1.11 – 3.48)	0.020
Mantle Cell	20	2.23 (1.15 – 4.31)	0.017
Peripheral T-cell	15	0.61 (0.21 – 1.76)	0.360
Others	38	1.50 (0.80 – 2.80)	0.204

Abbreviations: CI = confidence interval.

^aReference group

^b2 degrees of freedom

Table 7.

Multivariate analysis comparing survival among patients who underwent matched unrelated donor allogeneic hematopoietic stem cell transplant with myeloablative conditioning for NHL.

Variables:	N	Relative Risk (95% CI)	P-value
Age at transplant
≤20	46	1.00a	Pboverall = 0.025
21–40	105	1.12 (0.73 – 1.74)	0.596
≥40	105	1.72 (1.07 – 2.75)	0.025
Karnofsky performance status
≥90	162	1.00a
<90	94	1.38 (1.02 – 1.87)	0.037
Histology
Follicular	52	1.00a	Pcoverall = <0.001
DLCL/Immunoblastic	70	1.91 (1.23 – 2.99)	0.004
Lymphoblastic/Burkitts/Burkitt-like	61	2.25 (1.38 – 3.67)	0.001
Mantle Cell	20	2.16 (1.20 – 3.88)	0.010
Peripheral T-cell	15	0.44 (0.15 – 1.27)	0.128
Others	38	1.67 (0.99 – 2.80)	0.054
Disease status at transplant
Within 3 months post transplantd
Chemosensitive CR	67	1.00a	Pboverall = <0.001
Chemosensitive PR	93	1.70 (0.90 – 3.24)	0.103
Chemoresistant	96	2.99 (1.62 – 5.52)	<0.001

Abbreviations: CI = confidence interval.

^aReference group

^b2 degrees of freedom

^c4 degress of freedom

^dTime dependant covariates. The effect of disease status at transplant differs with the length of time after transplant. Patients transplanted with chemoresistant disease had no different death rates than patients transplanted with chemosensitive disease, when considering time period beyond 3 months after transplant. However, when considering the first 3 months after transplant patients transplanted with chemoresistant disease had higher rates of death.

Treatment Related Mortality

After adjustment for covariates the relative risk of TRM was 3.02 for those over the age of 40 (p<0.001) compared to patients aged 20 or younger. Karnofsky performance score (KPS) of less than 90 was also associated with a fifty percent increased risk for TRM (RR=1.43, p = 0.05). Compared to those with follicular lymphoma or those with peripheral T-cell lymphoma, patients with diffuse large cell, lymphoblastic/Burkitt or mantle cell lymphoma had significantly worse TRM.

Lymphoma Relapse/Progression

The risk for disease recurrence was increased fivefold (RR = 5.74, p= 0.002) for those undergoing transplant with refractory disease and fourfold (RR = 4.42, p=0.008) for those undergoing transplant in partial remission as compared to those undergoing transplant in complete remission. Lymphoblastic or Burkitt’s histology was also associated with increased disease recurrence (RR = 3.53, p=0.003).

Progression Free Survival and Treatment Failure

Progression free survival was adversely associated with lower performance status, aggressive disease histology and chemotherapy resistance. KPS <90 was associated with a higher risk of treatment failure (RR = 1.39, p=0.03). Patients with diffuse large cell, lymphoblastic/Burkitt’s/Burkitt like or mantle cell histology were at higher risk of treatment failure than those with follicular or peripheral T-cell lymphoma. Patients in partial remission at transplant (RR = 2.13, p= 0.01) or those with chemotherapy resistant disease (RR = 3.25, p<0.001) were at higher risk of treatment failure within 3 months of transplant.

Survival

Survival was superior for younger patients, those with better performance status, those with follicular or peripheral T-cell lymphoma and those with chemotherapy sensitive disease. Figure 2 shows survival curves by disease histology. Figure 3 illustrates the impact of KPS on survival and figure 4 the impact of disease stage.

Figure 2.

Overall Survival after unrelated donor allogeneic hematopoietic stem cell transplant for NHL by histology

Figure 3.

Overall Survival after unrelated donor allogeneic hematopoietic stem cell transplant for NHL by karnofsky performance status

Figure 4.

Overall Survival after unrelated donor allogeneic hematopoietic stem cell transplant for NHL by chemosensitivity

Causes of Death

Two hundred and ten patients died with twenty nine percent of the deaths (29%) attributable to progressive lymphoma. Organ failure (19%), infection (15%), pulmonary syndrome (13%) and GVHD (12%) were the other major causes of mortality. Relapse accounted for 39% of deaths in patients with lymphoblastic or Burkitt’s lymphoma (P<0.01 vs all others).

DISCUSSION

Multiple prospective and observational studies have demonstrated the curative potential of sibling transplantation in a variety of lymphoma subtypes and have established it as a reasonable alternative to autologous transplantation (8–14, 16, 17, 19–23, 37–39). Our analysis is the first large scale report on outcome of unrelated donor transplantation in patients with lymphoma. These patients were transplanted over a fourteen year period between 1991 and 2004. Many of the patients had chemotherapy refractory disease and many had a decreased performance score. Ninety percent engrafted, and twenty six percent of patients survived free of disease for up to five years. Thus unrelated donor transplantation can be considered a beneficial and curative procedure for a fraction of patients with NHL. Disconcertingly, the treatment related mortality was as high as 50% and was variably caused by organ toxicity, GVHD and infections (table 8). This is reminiscent of the very high treatment related mortality observed in early studies of HLA-identical transplantation for low grade lymphoma where small numbers of patients were reported from multiple institutions (40). Baseline patient and disease-related adverse risk factors should be considered when assessing candidacy for unrelated donor transplantation. Those risk factors included the chemotherapy refractory nature of the disease and a decreased performance status in many patients. Disease status (9, 16, 41, 42) and performance score (9, 40, 43–45) have previously been identified as predictors of outcome. With better patient selection it is to be expected that outcomes would be much improved. As in previous studies, older age and particularly age over 40 was another independent predictor for outcome, because of increased treatment related mortality associated with increasing age (8, 29, 40, 46). It is likely that older patients, in part because of co-morbidities, are less tolerant of the intensive conditioning regimens, and that they would benefit from reduced intensity conditioning as was recently shown by the Seattle group (42).

Table 8.

Causes of death among patients who underwent matched unrelated donor allogeneic hematopoietic stem cell transplant with myeloablative conditioning for NHL.

Causes of death	N eval	N (%)
Number of patients	210
Primary disease		60 (29)
GVHD		25 (12)
Pulmonary syndrome		27 (13)
Infection		32 (15)
Organ Failure		41 (19)
Others*		25 (12)

^*Others include (N=25)

1. New malignancy (n=3)

2. Hemorrhage (n=11)

3. Vascular (n=1)

4. Acute cardiac infarct (n=1)

5. Failure to thrive (n=1)

6. TTP (n=1)

7. Metabolic abnormalities (n=1)

8. Non-bacterial thrombotic (n=1)

9. Status epilepticus (n=1)

10. Other, not specified (n=4)

The marked influence of disease histology on TRM and on disease recurrence is also consistent with numerous prior observations. Patients with follicular lymphoma and those with peripheral T-cell lymphoma had the lowest risk of treatment related mortality and of disease recurrence. The low recurrence rate in follicular lymphoma is also consistent with previous observations. A relatively low rate of treatment related mortality has by many been attributed to the use of reduced intensity conditioning (11, 47). In registry analysis however, the treatment related mortality of myeloablative transplantation for follicular lymphoma has also decreased over time and is not different than that associated with reduced intensity conditioning (9, 48, 49). The relatively good outcome in the small group of fifteen PTCL patients, twelve of whom had recurrent or refractory disease, is gratifying and adds support to emerging data on the role of allogeneic transplantation in this otherwise difficult to treat disorder (21, 50). Having aggressive B-cell lymphomas was a independent risk factor for disease recurrence, and also for treatment related mortality. While disease relapse can be attributed to disease biology, it is more difficult to understand how disease histology could relate directly to TRM. We speculate that patients with aggressive B-cell lymphomas either because of disease biology or more intensive prior treatment present to transplant with a more pro-inflammatory host milieu, markers of which have been repeatedly shown to be associated with TRM (51, 52). Donor matching is an increasingly important determinant of outcome of unrelated donor transplantation, particularly in patients with a favorable disease stages(2). Since so many patients in the current study had advanced disease, it is not surprising that no impact of HLA typing on outcome could be detected.

In summary, unrelated donor HCT induced long term survival in approximately 25% of patients with lymphoma transplanted between 1991 and 2004. The outcomes are poor for those with refractory disease and decreased performance status, but encouraging in patients with follicular lymphoma and T-cell lymphoma. It is likely that with better patient and donor selection, improved supportive care, and changes in conditioning and GVHD prophylaxis, current results are superior, as was shown in single institution studies and in registry studies of sibling transplant for follicular lymphoma (42, 49).Unrelated donor transplantation should be considered a treatment alternative for patients with high risk, recurrent or refractory lymphoma, particularly for those with good performance status and chemotherapy sensitive disease.

Table 5.

Multivariate analysis comparing progression/relapse among patients who underwent matched unrelated donor allogeneic hematopoietic stem cell transplant with myeloablative conditioning for NHL.

Variables:	N	Relative Risk (95% CI)	P-value
Disease status at transplant
Within 3 months post transplanta
Chemosensitive CR	67	1.00b	Pcoverall = 0.008
Chemosensitive PR	93	4.42 (1.46 – 13.34)	0.008
Chemoresistant	96	5.74 (1.90 – 17.36)	0.002
Histology
Follicular	52	1.00b	Pcoverall = 0.020
DLCL/Immunoblastic	70	1.69 (0.72 – 4.00)	0.231
Lymphoblastic/Burkitts/Burkitt-like	61	3.53 (1.53 – 8.14)	0.003
Mantle Cell	20	2.46 (0.74 – 8.21)	0.143
Peripheral T-cell	15	0.37 (0.05 – 2.96)	0.349
Others	38	2.49 (1.02 – 6.10)	0.046

Abbreviations: CI = confidence interval.

^aTime dependant covariates. The effect of disease status at transplant differs with the length of time after transplant. Patients transplanted with chemoresistant disease had no different progression/relapse rates than patients transplanted with chemosensitive disease, when considering time period beyond 3 months after transplant. However, when considering the first 3 months after transplant patients transplanted with chemoresistant disease had higher rates of progression/relapse.

^bReference group

^c2 degrees of freedom

Table 6.

Multivariate analysis comparing progression-free survival among patients who underwent matched unrelated donor allogeneic hematopoietic stem cell transplant with myeloablative conditioning for NHL.

Variables:	N	Relative Risk (95% CI)	P-value
Karnofsky performance status
≥90	162	1.00a
<90	94	1.39 (1.03 – 1.87)	0.031
Histology
Follicular	52	1.00a	Pboverall <0.001
DLCL/Immunoblastic	70	1.56 (1.02 – 2.39)	0.041
Lymphoblastic/Burkitts/Burkitt-like	61	1.92 (1.23 – 2.99)	0.004
Mantle Cell	20	2.62 (1.48 – 4.66)	0.001
Peripheral T-cell	15	0.43 (0.17 – 1.11)	0.083
Others	38	1.41 (0.86 – 2.32)	0.170
Disease status at transplant
Within 3 months post transplantc
Chemosensitive CR	67	1.00a	Pdoverall = <0.001
Chemosensitive PR	93	2.13 (1.20 – 3.78)	0.010
Chemoresistant	96	3.25 (1.86 – 5.70)	<0.001

Abbreviations: CI = confidence interval.

^aReference group

^b4 degress of freedom

^cTime dependant covariates. The effect of disease status at transplant differs with the length of time after transplant. Patients transplanted with chemoresistant disease had no different progression/relapse or death rates than patients transplanted with chemosensitive disease, when considering time period beyond 3 months after transplant. However, when considering the first 3 months after transplant patients transplanted with chemoresistant disease had higher rates of progression/relapse or death.

^d2 degrees of freedom