Outcomes and a New Prognostic Model in Posttransplantation Lymphoproliferative Disorders (PTLD): Intent-to-Treat (ITT) Analysis of 81 Patients in the Post-Rituximab Era.

Evens, Andrew M.; David, Kevin A.; Helenowski, Irene; Kircher, Sheetal M.; Mauro, Lauren; Gimelfarb, Alla; Van Besien, Koen; Hattersley, Elise; Nelson, Beverly P; Mehta, Jayesh; Gordon, Leo I.; Shammo, Jamile M.; Smith, Scott E.; Smith, Sonali

doi:10.1182/blood.V112.11.1776.1776

Abstract

The outcome of PTLD has a reported approximate 3-year (yr) overall survival (OS) rate of 35%–40% in the pre-RTX era (

JCO 2001; 19:772

). The impact, if any, of RTX on the prognosis or outcome of PTLD is not known. We retrospectively studied the clinical features, treatment, outcomes (ITT), and prognostic factors among a large multi-center cohort of solid organ transplantation (SOT)-related PTLD cases over a 10-yr period in the post-RTX era (1/98–2/08) at 4 Chicago academic institutions. Prognostic factors were evaluated in Cox proportional hazards regression for indicators of progression-free survival (PFS) and OS. 81 pts were identified with a median age at PTLD diagnosis (dx) 48 yrs (range, 20–72) and median time from SOT to PTLD dx 42 months (mo) (range, 1-216 mo). EBV+ PTLD pts more commonly were treated by concurrent reduction of immune suppression (median reduction 75%) with single agent RTX whereas EBV-neg pts more frequently received RTX/chemotherapy with median immune suppression reduction of 90%. On ITT (median follow-up 36 mo), the 3-yr PFS for all pts was 58% and 3-yr OS was 62%. Of note, 13/81 (16%) pts died ≤6 weeks from PTLD dx, primarily due to progressive disease. Univariate analysis found several factors highly predictive of outcome (Table 1):

performance status (PS),
serum albumin,
>1 extranodal (EN) site,
bone marrow (BM) involvement,
CNS disease, and
RTX as part of initial therapy (alone or combined with chemotherapy).

Neither EBV status nor time to PTLD predicted outcome.

Table 1. PTLD characteristics and univariate predictors of EFS and OS (n=81).

Variable (at dx)		Pt characteristics		PFS		OS
		No.	%	HR*	P	HR*	P
*Prognostic factors were tested for association with survival by proportional hazards analysis. Hazard ratios (HRs) >1 indicate a poor prognosis, while HRs <1 indicate a good prognosis.
**Analysis was monomorphic PTLD (n=55) vs polymorphic and reactive (n=26).
***Analysis combined rituximab alone and rituximab/chemotherapy-treated pts (n=60) vs all other therapy (n=21).
n/a = not available.
Table 2 shows 3-yr survival rates based on these variables. On multivariate analysis, 4 factors remained significant:
1) RTX (EFS HR 0.26 (CI 0.11–0.60), p=0.002; OS HR 0.23 (CI 0.090.56), p=0.001),
2) >1 EN site (EFS HR 2.15 (CI 1.04–4.42), p=0.04; OS HR 2.14 (CI 1.01–4.54), p=0.047),
3) low albumin (EFS HR 3.19 (CI 1.17–8.72), p=0.02; OS HR 4.73 (CI 1.39–16.16), p=0.01), and
4) CNS (EFS HR 4.48 (CI 1.80–11.17), p=0.01; OS HR 3.90 (CI 1.54–9.88), p=0.004).
A prognostic model based on 3 factors: >1 EN site, albumin, and CNS disease was constructed (EFS p=0.008, OS=p=0.006), suggesting that pts with differential outcomes can be identified. In summary, we show that the introduction of RTX appears to have improved the outcome of PTLD compared to published data. However, a significant minority of pts still experience short survival. Clinical features at PTLD dx can predict outcome and should be incorporated into future therapeutic strategies.
	Kidney	37	46
	Kidney-pancreas	10	12
SOT	Pancreas	4	5	0.74	0.38	0.71	0.36
	Liver	17	21
	Heart	8	10
	Lung	5	6
Time to PTLD	Early (<1 yr)	30	37	1.02	0.95	0.91	0.81
	Late (> 1 yr)	51	63
	Monomorphic	55	68
Histology**	Polymorphic	22	27	1.61	0.27	2.18	0.11
	Plasmacytic/reactive	4	5
EBV + (tumor)	Yes	39	48
	No	30	37	1.09	0.82	0.82	0.64
	n/a	12	15
PS	2–4	25	31	2.87	0.002	3.31	0.001
	0–1	56	69
CNS involvement	Yes	10	12	2.57	0.03	2.47	0.050
	No	71	88
BM involvement	Yes	12	15
	No	58	72	3.30	0.003	3.48	0.003
	n/a	11	13
EN > 1	Yes	29	36
	No	50	62	2.01	0.04	2.19	0.03
	n/a	2	2
High LDH	Yes	49	60
	No	27	33	1.64	0.21	2.00	0.11
	n/a	5	7
	Yes	56	69
Low serum albumin	No	24	30	3.02	0.02	4.72	0.01
	n/a	1	1
IPI	2–4	47	58
	0–2	33	41	2.04	0.053	1.76	0.16
	n/a	1	1
	Immune suppression alone	7	9
Treatment***	Radiation or surgery alone	2	2
	Chemotherapy alone	12	15	0.28	0.003	0.30	0.001
	Rituximab alone	26	32
	Rituximab/chemotherapy	34	42

Variable (at dx)		Pt characteristics		PFS		OS
		No.	%	HR*	P	HR*	P
*Prognostic factors were tested for association with survival by proportional hazards analysis. Hazard ratios (HRs) >1 indicate a poor prognosis, while HRs <1 indicate a good prognosis.
**Analysis was monomorphic PTLD (n=55) vs polymorphic and reactive (n=26).
***Analysis combined rituximab alone and rituximab/chemotherapy-treated pts (n=60) vs all other therapy (n=21).
n/a = not available.
Table 2 shows 3-yr survival rates based on these variables. On multivariate analysis, 4 factors remained significant:
1) RTX (EFS HR 0.26 (CI 0.11–0.60), p=0.002; OS HR 0.23 (CI 0.090.56), p=0.001),
2) >1 EN site (EFS HR 2.15 (CI 1.04–4.42), p=0.04; OS HR 2.14 (CI 1.01–4.54), p=0.047),
3) low albumin (EFS HR 3.19 (CI 1.17–8.72), p=0.02; OS HR 4.73 (CI 1.39–16.16), p=0.01), and
4) CNS (EFS HR 4.48 (CI 1.80–11.17), p=0.01; OS HR 3.90 (CI 1.54–9.88), p=0.004).
A prognostic model based on 3 factors: >1 EN site, albumin, and CNS disease was constructed (EFS p=0.008, OS=p=0.006), suggesting that pts with differential outcomes can be identified. In summary, we show that the introduction of RTX appears to have improved the outcome of PTLD compared to published data. However, a significant minority of pts still experience short survival. Clinical features at PTLD dx can predict outcome and should be incorporated into future therapeutic strategies.
	Kidney	37	46
	Kidney-pancreas	10	12
SOT	Pancreas	4	5	0.74	0.38	0.71	0.36
	Liver	17	21
	Heart	8	10
	Lung	5	6
Time to PTLD	Early (<1 yr)	30	37	1.02	0.95	0.91	0.81
	Late (> 1 yr)	51	63
	Monomorphic	55	68
Histology**	Polymorphic	22	27	1.61	0.27	2.18	0.11
	Plasmacytic/reactive	4	5
EBV + (tumor)	Yes	39	48
	No	30	37	1.09	0.82	0.82	0.64
	n/a	12	15
PS	2–4	25	31	2.87	0.002	3.31	0.001
	0–1	56	69
CNS involvement	Yes	10	12	2.57	0.03	2.47	0.050
	No	71	88
BM involvement	Yes	12	15
	No	58	72	3.30	0.003	3.48	0.003
	n/a	11	13
EN > 1	Yes	29	36
	No	50	62	2.01	0.04	2.19	0.03
	n/a	2	2
High LDH	Yes	49	60
	No	27	33	1.64	0.21	2.00	0.11
	n/a	5	7
	Yes	56	69
Low serum albumin	No	24	30	3.02	0.02	4.72	0.01
	n/a	1	1
IPI	2–4	47	58
	0–2	33	41	2.04	0.053	1.76	0.16
	n/a	1	1
	Immune suppression alone	7	9
Treatment***	Radiation or surgery alone	2	2
	Chemotherapy alone	12	15	0.28	0.003	0.30	0.001
	Rituximab alone	26	32
	Rituximab/chemotherapy	34	42

Table 2. Three-year survival (Kaplan-Meier) based on prognostic variables (univariate).

Variable	3-year EFS*	3-year OS*
*3-year PFS and OS by Kaplan-Meier analysis are reported (and their 95% CIs).
PS	p < 0.0001	p < 0.0001
0–1	65% (50%, 77%)	72% (57%, 82%)
2–4	36% (18%, 54%)	35% (15%, 56%)
Albumin	p = 0.002	p = 0.0004
Normal	75% (49%, 89%)	85% (59%, 95%)
Low	49% (34%, 61%)	51% (37%, 64%)
CNS	p = 0.001	p = 0.004
No	60% (47%, 71%)	64% (51%, 74%)
Yes	30% (7%, 58%)	40% (12%, 67%)
Received rituximab	p = 0.0002	p = 0.001
No	21% (7%, 42%)	32% (14%, 79%)
Yes	69% (55%, 79%)	71% (57%, 81%)
BM involvement	p = 0.005	p = 0.005
No	64% (49%, 75%)	70% (56%, 80%)
Yes	18% (3%, 44%)	18% (3%, 44%)
> 1 EN site	p = 0.01	p = 0.008
No	64% (48%, 76%)	72% (57%, 82%)
Yes	39% (22%, 58%)	40% (20%, 58%)

Variable	3-year EFS*	3-year OS*
*3-year PFS and OS by Kaplan-Meier analysis are reported (and their 95% CIs).
PS	p < 0.0001	p < 0.0001
0–1	65% (50%, 77%)	72% (57%, 82%)
2–4	36% (18%, 54%)	35% (15%, 56%)
Albumin	p = 0.002	p = 0.0004
Normal	75% (49%, 89%)	85% (59%, 95%)
Low	49% (34%, 61%)	51% (37%, 64%)
CNS	p = 0.001	p = 0.004
No	60% (47%, 71%)	64% (51%, 74%)
Yes	30% (7%, 58%)	40% (12%, 67%)
Received rituximab	p = 0.0002	p = 0.001
No	21% (7%, 42%)	32% (14%, 79%)
Yes	69% (55%, 79%)	71% (57%, 81%)
BM involvement	p = 0.005	p = 0.005
No	64% (49%, 75%)	70% (56%, 80%)
Yes	18% (3%, 44%)	18% (3%, 44%)
> 1 EN site	p = 0.01	p = 0.008
No	64% (48%, 76%)	72% (57%, 82%)
Yes	39% (22%, 58%)	40% (20%, 58%)

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author

2008, The American Society of Hematology

2008

Sign in via your Institution

Outcomes and a New Prognostic Model in Posttransplantation Lymphoproliferative Disorders (PTLD): Intent-to-Treat (ITT) Analysis of 81 Patients in the Post-Rituximab Era.

Abstract

Author notes

Contents

Data & Figures

Supplemental data

References

Cited By

Email alerts

ASH Publications

American Society of Hematology

Outcomes and a New Prognostic Model in Posttransplantation Lymphoproliferative Disorders (PTLD): Intent-to-Treat (ITT) Analysis of 81 Patients in the Post-Rituximab Era.

Abstract

Author notes

Contents

Data & Figures

Supplemental data

References

Related

Related

Cited By

Email alerts

ASH Publications

American Society of Hematology

This Feature Is Available To Subscribers Only