Abstract
Background. PTLD represent one of the most serious hematological complications of solid-organ transplantation. Our objective was to determine clinical and immunohistopathological prognostic factors for overall survival (OS) in adult patients with PTLD.
Methods. Eighty-four patients diagnosed with PTLD between1980 and 2004 at University of Pittsburgh Medical Center were identified for this study. Immunohistochemistry staining was performed on tumor tissue at the time of diagnosis for the following proteins: BCL2, BCL6, cMYC, p53, phosphorylated ribosomal S6kinase, HSP90 and NFk-B. Univariate associations between individual clinical and immunohistopathological features and OS were compared using Cox proportional hazard models. Survival curves were generated via the method of Kaplan-Meier.
Results. The median age at the time of diagnosis of PTLD was 53 years (range 18–77). The median time from transplant to PTLD was 32 months (range 0.87–288). Twenty-nine (34%) patients presented with early PTLD (within 1 year of transplant). The tumors were EBV positive by in situ hybridization in 64% of the patients, and CD20 positive in 60%. The majority of the patients presented with extranodal disease (73%). Twenty-two (27%) patients had involvement of the grafted organ. Reduction of immunosuppression was performed in 77% of the cases, and rituximab was given in 27% of the patients. Lung transplant was significantly related to grafted organ involvement (p=0.035). Lung and multiorgan transplants were associated with a higher risk for early PTLD (p=0.0033 and p=0.054 respectively), while renal transplant was associated with risk for late PTLD (p=0.0011). The median survival for all patients was 20.8 months, 95% CI: (7.4–77.6). On univariate analysis for OS, the following poor prognostic factors were identified: age of transplant >60 yr (p=0.02), multiorgan transplant (p=0.02), ECOG>2 (p<0.0001), grafted organ involvement and extranodal disease at the time of diagnosis (p<0.011), early PTLD (p<0.0001), and elevated white blood cell count (WBC) 1 month prior to diagnosis (p=0.01). Good prognostic factors included ECOG of 1 (p=0.0004), late PTLD (p=0.002), early stage at diagnosis (stage I and II, p=0.005), monomorphic disease (p=0.0034), CD20 positive (p=0.033), initial immunosuppression reduction (p=0.05) and rituximab use (p=0.047). We analyzed whether the presence of BCL2, BCL6, cMYC, p53, pS6, HSP90 and NFk-B correlated with prognosis, and demonstrated that only BCL2 strongly correlated with prognosis (p=0.0020). PTLD-related survival strongly correlated with the intensity of BCL2 staining. A final clinico-pathological multivariable model for survival in PTLD was constructed using three factors: ECOG >2, elevated WBC one month prior diagnosis and BCL2 overexpression at the time of diagnosis. Patients with these three risk factors had median survival of 10 days (95% CI: 0–41), in comparison with patients with none of these risk factors had median survival of 1,414 days.
Conclusion: We developed a prognostic model based on clinical and pathological markers. Patients who had all 3 factors ECOG >2, elevated WBC one month prior diagnosis, and BCL2 overexpression in tumor tissue had worse median survival than the patients with none of these three factors.
Disclosure: No relevant conflicts of interest to declare.
Author notes
Corresponding author
