Abstract 113


DLBCL and HL represent highly curable lymphoid malignancies. Patients (pts) whose lymphoma is refractory to or relapses following initial therapy pose a significant therapeutic challenge. The goal of therapy is to proceed to a non-cross-resistant salvage regimen followed by high dose chemotherapy (HDC) and stem cell transplantation (SCT) for transplant eligible pts. The optimal choice of salvage therapy remains unknown. The combination of gemcitabine, dexamethasone and cisplatin (GDP) has been shown in phase II studies to induce high response rates with minimal toxicity (Baetz T, Ann Oncol, 2003; Crump M, Cancer, 2004). Based on these promising results, British Columbia Cancer Agency (BCCA) policy has recommended GDP as the preferred salvage regimen for pts with relapsed/refractory DLBCL and HL since 2002.

Patients and Methods:

We conducted a retrospective analysis using the BCCA Lymphoid Cancer Database and included all pts with relapsed/refractory DLBCL and HL who received GDP as salvage therapy between September 2002 and June 2010. Pts were treated with gemcitabine 1000 mg/m2 IV day 1,8; dexamethasone 40 mg PO days 1–4 and cisplatin 75 mg/m2 IV day 1, administered at 3 week intervals (2-3 cycles for transplant eligible patients and up to 6 cycles for non-transplant candidates). Primary endpoints were response rate, PFS (defined as the interval from the beginning of GDP to first progression, relapse or death from any cause) and OS.


235 pts treated with GDP were identified; 152 and 83 pts with relapsed/refractory DLBCL and HL, respectively.

Clinical characteristics at time of diagnosis for patients with DLBCL were: 68% male, 65% stage III/IV, 42% bulky disease ≥ 10 cm, 43% B-symptoms, 59% IPI 0–2, 41% IPI 3–5. Median age at time of GDP was 57 y (range 20–79 y). 57 pts (37%) had primary refractory disease to first-line R-CHOP; 144 (95%) were treated with GDP at first relapse/progression; median time from diagnosis to relapse after R-CHOP (excluding primary refractory pts) was 21 m (range 7–139 m). 30 pts (20%) received rituximab with GDP. Detailed radiologic response assessment following GDP(+/−R) was available for 82% pts with response rates as follows: 16% CR/CRu, 33% PR, 17% SD, 34% PD. 9 pts (6%) underwent HDC followed by allogeneic SCT and 57 pts (38%) underwent HDC followed by autologous SCT. With median follow-up of 24 m from start of GDP (range 0–84 m), 51 pts (34%) were alive and 101 pts (66%) have died (99 from lymphoma, 1 treatment toxicity during allogeneic SCT, 1 unrelated cause). 2-y PFS and OS were 21% and 28%, respectively. The 2-y PFS and OS for the subset of patients who underwent HDC/SCT were 36% and 47%, respectively.

Clinical characteristics at diagnosis for the 83 pts with relapsed/refractory HL were: 55% male, 59% stage III/IV, 39% bulky disease ≥ 10 cm, 55% B-symptoms, 80% nodular sclerosis, 4% mixed cellularity, 2% nodular lymphocyte predominant, 2% lymphocyte depleted and 12% HL NOS. IPS variables were retrievable on 66% patients: 82% IPS ≤ 3 and 18% IPS ≥ 4. Median age at time of GDP was 31 y (range 17–73 y). 30 pts (36%) had primary refractory HL and 73 (88%) received GDP at first relapse/progression. Median time from diagnosis to relapse following ABVD-like therapy (excluding primary refractory pts) was 20 m (range 9–186 m). Detailed radiologic response assessment following GDP was available in 67% pts with response rates as follows: 7% CR/CRu, 64% PR, 13% SD, 16% PD. In total, 1 pt underwent HDC followed by allogeneic SCT and 69 pts (83%) proceeded to HDC and autologous SCT. With a median follow-up of 30 m from start of GDP (range 0–86 m), 70 pts (84%) were alive and 13 (16%) have died (all from HL). 2-y PFS and OS were 58% and 85%, respectively, and for the subset of pts who underwent HDC/SCT were 57% and 86%, respectively. Hospitalization rates due to complications during GDP were higher in patients with DLBCL than HL (20% vs 7%), likely reflecting differences in age and co-morbidities between the 2 cohorts. No failures of stem cell mobilization were recorded and the only toxic death was a consequence of HDC and allogeneic SCT.


GDP is an effective and well-tolerated out-patient salvage regimen for relapsed/refractory DLBCL and HL. Outcomes appear to be comparable to those reported with more aggressive regimens. Results from an ongoing Canadian prospective trial comparing R-GDP to R-DHAP will help clarify the role of GDP in the treatment of relapsed/refractory DLBCL.


Connors:Hoffmann-La Roche: Research Funding.

This icon denotes an abstract that is clinically relevant.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution