Treatment results for extranodal natural killer/T-cell lymphoma (ENKTL) have historically been poor. With CHOP and CHOP-like chemotherapy, median survival is only 2-7.8 months with 5-year overall survival (OS) of ~30%.r The poor outcome is partly due to NK-cell expression of high levels of multi-drug resistant (MDR) P-glycoprotein, so that anthracycline-containing regimens are ineffective. Standard of care for advanced stage ENKTL has been SMILE (dexamethasone, methotrexate, ifosfamide, asparaginase and etoposide), based on a phase II study of 38 patientsr and a phase II multicentre study in a non-trial setting of 87 patients.r These studies have shown an overall response rate (ORR) of 79-81% and complete response (CR) of 50-56% after 2-3 cycles. 5-year OS was 50%, and 4-year disease-free survival (DFS) was 64%. SMILE has significant toxicity, with grade III/IV neutropenia in 72.7% and a treatment-related mortality (TRM) of 7%.
The DDGP protocol was developed due to toxicity issues related to SMILE, as well as the high rates of hypersensitivity reactions to L-asparaginase. The expert reference panel supported publication of the protocol on the basis of the information summarised below. The committee was most strongly influenced by the randomised controlled, open-label, multicentre study published in 2016 by Li et al. comparing DDGP and SMILE in 42 advanced stage, newly diagnosed patients.r An abstract of the follow-up to this study was presented at ASH 2019.r A total of 80 patients with newly diagnosed advanced stage ENKTL with an ECOG of 0-2 and aged between 14 and 70 years old were recruited between 2011 and 2019. Patients were randomised 1:1 to SMILE or DDGP, and given up to 6 cycles unless disease progression, unacceptable toxicity or patient choice.
Zhang et al. published a retrospective study of 80 patients in 2016, with 60% of patients being newly diagnosed, 11% in first relapse, and 29% in a primary refractory disease.r Of the newly diagnosed, 42% were stage I-II and the remainder were stage III-IV. More recently, DDGP has also been trialled in early-stage disease. A randomised, controlled, open-label, multicentre study compared DDGP combined with radiation therapy (RT) vs RT alone in newly diagnosed stage I-II NKTL.r Published in 2021, the study reported on 65 patients, with 35 patients receiving RT alone, and 30 treated with DDGP and RT. Another prospective cohort study from 2011 to 2016 examined effects of DDGP followed by RT (DDGP+RT) versus VIPD followed by RT (VIPD+RT) in 40 newly diagnosed stage I-II ENKTL patients.r
A real-world retrospective analysis of 376 patients on asparaginase-based vs non-asparaginase-based chemotherapy combined with RT in the treatment of early-stage ENKTCL demonstrated a significantly improved 5-year OS (84.5% vs 73.2%) and distant metastasis-free survival (84.4% vs 74.5%) with the asparaginase-based regimen in intermediate- to high-risk patients.r Out of the asparaginase-based group, 13 patients had DDGP overall, and 8 were in the intermediate- to high-risk group. Other asparaginase-based regimens included were GELOX, P-GEMOX and GDP-L.
Source |
Study & Year Published |
Supports Use |
Is the dose and regimen consistent with the protocol? |
Comments |
Phase II trials |
Li et al. 2016r |
Yes |
Yes |
Newly diagnosed, advanced stage. |
Wang et al. 2019r (Abstract) |
Yes |
Yes |
Newly diagnosed, advanced stage. |
Zhang et al. 2021r |
Yes |
Yes |
Newly diagnosed, early stage |
Observational studies |
Zhang et al. 2016r |
Yes |
Yes |
Newly diagnosed, relapsed and primary refractory. 25% stage I-II. |
Wang et al. 2021r |
Yes |
Yes |
Relapsed/refractory |
Zhang et al. 2022r |
Yes |
Yes |
Newly diagnosed, early stage |
Zheng et al. 2021r |
Yes |
N/A |
Newly diagnosed, early stage |
Guidelines |
Date published/revised |
Supports Use |
Is the dose and regimen consistent with the protocol? |
Comments |
NCCN |
V 1.2022 |
Yes |
N/A |
Consider clinical trial, DDGP, modified SMILE or P-GEMOX for induction in advanced disease |
BCCA |
N/A |
N/A |
N/A |
N/A |
CCO |
Jan 2022 |
Yes |
Yes |
N/A |
Efficacy
For advanced stage, newly diagnosed ENKTL, Li et al. reported an overall response rate (ORR) of 95% vs 67% for SMILE; and a complete response (CR) of 71% vs 29% for SMILE. 1-year (overall survival) OS was 90% vs 57% for DDGP vs SMILE and progression-free survival (PFS) 86% for DDGP vs 38% for SMILE. 2-year OS was 74% vs 45% for DDGP and SMILE, respectively. The response rates for SMILE were significantly lower than those reported in previous clinical trials.r In the follow-up study, the ORR was significantly higher in DDGP vs SMILE (90% vs 60%), but there was no significant difference in CR (67.5% for DDGP vs 47.5% for SMILE). The 3-year PFS was 56.6% for DDGP and 41.8% for SMILE; 5-year OS 74.3% for DDGP and 51.7% for SMILE.r
Figure 1.Overall survival (A) and progression-free survival (B) curvesr

© AACR 2016
The 2016 retrospective study by Zhang et al. reported an ORR of 91.3% and CR of 60%. The 1-year OS and PFS were 88.6% and 86.1%, and 2-year OS and PFS were 87% and 81%.r Another retrospective study of 54 patients with relapsed/refractory disease treated with either DDGP or SMILE in 2021 demonstrated that the 31 patients who received DDGP had a higher CR rate (61.3% vs 30.4%) and 5-year PFS (45.4% vs 27.6%) and 5-year OS (50.1% vs 32.5%) compared to the SMILE group.r There was no significant difference in the ORR (83.9% vs 60.9%) between the two groups.
The 2021 study by Zhang et al. comparing DDGP+RT versus RT alone reported the combined modality group having superior CR (73.3% vs 48.6%) and ORR (83.3% vs 60%). 5-year PFS (82.9% vs 56.5%) and OS (85.7% vs 60.4%) were also significantly better in the combined modality group.r Another prospective cohort study comparing DDGP+RT and VIPD+RT reported better CR rate (85% vs 50%) and OR rate (95% vs 65%) in the DDGP+RT arm .r Whilst 5-year PFS was better in the DDGP+RT group (83.3% vs 44.4%), there was no significant difference in the 5-year OS between the two groups (83.0% vs 72.1%).
Toxicity
Li et al. reported significantly fewer adverse events in the DDGP arm; particularly lower rates of leukopenia, neutropenia, hypersensitivity reactions and gastrointestinal side effects.r DDGP had higher rates of anaemia and coagulation abnormalities. Grade 3 or 4 leukopenia occurred in 13 patients (54%) with DDGP, compared to 19 (90%) with SMILE. Grade 3 or 4 anaemia occurred in 11 (52%) and thrombocytopenia in 13 (62%). There were no documented cases of allergic reactions with DDGP, but they occurred in 7 patients (33%) in the SMILE group. There were no reported deaths.
Adverse effects between DDGP and SMILE groupsr

© AACR 2016
In the expanded cohort, grade 3 or 4 haematologic toxicity occurred in 25 patients (62.5%) compared to 34 (85%) in the SMILE group.r Grade 3 or 4 neutropenia occurred in 26 patients (65%). Non-haematologic toxicity including elevated transaminases, mucositis or allergic reactions did not occur with DDGP in this study. A ~10% treatment-related mortality was reported for DDGP, compared to 17.5% with SMILE, both significantly higher compared to previous studies. These were mainly due to infection and haemorrhage in the SMILE group.
Wang et al. reported no episodes of allergic reaction or pancreatitis. 49% had prolonged APTT and 41% hypofibrinogenaemia.r Grade 3 - 4 neutropenia occurred in 65% of patients, grade 4 thrombocytopenia in 35% and grade 4 anaemia in 5%. There were no treatment-related deaths. If there was a grade 4 toxicity, doses were reduced not more than 20% for subsequent cycles.
For early stage disease, reported adverse events of the DDGP+RT were as follows: grade 3 or 4 myelosuppression in 20-25%, grade 3 or 4 liver dysfunction in 3.3%, grade 3 or 4 gastrointestinal reactions in 16.7%, grade 3 or 4 alopecia in 16.7%.r There were no grade 3 or 4 coagulation abnormalities. It is noted that the RT only group was biased towards having more patient with a lower ECOG. Another study examining combined modality treatment noted the DDGP+RT group were more prone to grade I-II clotting dysfunction compared to the VIPD+RT group.r