Anti-PD1 agents as consolidation in relapsed/refractory classical Hodgkin lymphoma

Classical Hodgkin lymphoma (HL) is a highly curable disease, however, up to 30% of patients with advanced disease relapse. In these patients, salvage chemotherapy and autologous stem cell transplant (ASCT) can induce long-term remissions. However, those patients relapsing after ASCT have poor outcomes with overall survival (OS) ranging from 10.5–27.6 months.¹

HL is uniquely vulnerable to PD-1 blockade and as such, anti-PD1 antibodies such pembrolizumab and nivolumab are being investigated in HL. In recent studies, checkpoint blockade with anti-PD1 antibodies has proved efficacious in relapsed/refractory (R/R) HL, with overall response rates (ORRs) of approximately 70% and complete responses (CRs) of up to 20%. In 2016 and 2017, the United States (US) Food & Drug Administration (FDA) approved nivolumab and pembrolizumab, respectively, in this setting.¹

However, questions remain on the long-term efficacy and optimal management of R/R HL, including duration of treatment and consolidation strategies following salvage chemotherapy and ASCT. The efficacy of pembrolizumab as consolidation post-ASCT was recently investigated by Philippe Armand, Dana-Farber Cancer Institute, Boston, US, and colleagues, in an article published in Blood.² Another study investigating consolidation strategies was published in the European Journal of Cancer by, Guillaume Manson, University Hospital of Rennes, Rennes, FR, and colleagues, who investigated the use of nivolumab, with and without allo-HSCT.¹

Pembrolizumab as consolidation post-ASCT²

The study by Armand et al., was a phase II (NCT02362997), multicohort, open-label study in the US investigating pembrolizumab as consolidation post-ASCT. The authors’ hypothesis was that pembrolizumab would improve progression-free survival (PFS) at 18 months post-ASCT from 60% to 80%. Additionally, the treatment would be considered promising if 22 of the 30 patients treated were alive and progression-free at 18 months.

Patient characteristics and study design

• Patients (N= 30) with R/R cHL who relapsed after or were refractory to frontline therapy
  • Patients must have had prior ASCT and must have achieved a partial response (PR) or complete metabolic response to salvage therapy prior to ASCT
  • No more than three prior lines of therapy (excluding ASCT)
• Intravenous (IV) pembrolizumab (200mg) was given every three weeks, up to eight cycles, starting within 21 days post-ASCT
  • No dose modifications were permitted but could be delayed for up to 12 weeks due to toxicity
• Primary endpoint: PFS at 18 months post-ASCT
• Secondary endpoints: safety, 18-month OS, PFS and OS in high-risk patients not in CR prior to ASCT and response to pembrolizumab in patients with measurable disease post-ASCT
• Patient characteristics:
  • Median age: 33 (20–69) years
  • High-risk: 90%
  • Primary refractory disease: 57% (n= 17)
  • Disease status at study entry (PR vs CR): 7% (n= 2) vs 93% (n= 28)
  • Most patients (77%) received doxorubicin, bleomycin, vinblastine and dacarbazine (ABVD) as frontline therapy

Safety

• Completion of 8 cycles: 77% (n= 23)
• Discontinuation due to:
  • Patient choice: n= 2
  • Toxicity: n= 4
  • Progressive disease (PD) prior to cycle eight: n= 1

Table 1. Adverse events (AEs) reported on study

Type of AE

N (%)

≥1 Grade ≥2 AE

24 (80%)

≥1 Grade ≥3 AE

9 (30%)

Treatment-related AE (TRAE)

 

≥1 Grade ≥2 TRAE

15 (50%)

≥1 Grade ≥3 TRAE

8 (27%)

In total, 39 immune related AEs (irAEs) occurred in 13 patients (43%). The most frequent grade ≥ 2 events were pneumonitis, cough or dyspnea.

Efficacy

• CR rate at 12 months: 87% (n= 26)
• PFS at 18 months:
  • Relapse: 17% at a median of 6 months (3–18) post-ASCT
  • CR: 83%
  • The study therefore met the primary endpoint of > 80% PFS
• OS at 19 months: 100%

 Nivolumab consolidation with or without allo-HSCT¹

The study by Manson et al., retrospectively analyzed 78 adult patients with R/R HL who were treated with nivolumab in the French Early Access Program (EAP) and compared their outcomes based on consolidation with or without allo-HSCT.¹

Patient characteristics and study design

The French EAP included patients who had received three lines of chemotherapy including brentuximab vedotin and ASCT, or four lines of chemotherapy if the patient was ineligible for transplant.¹

• Median age: 37 (18–77) years
• Stage III/IV disease: 77.5%
• Median prior lines of treatment: 6 (2–13)
• Prior transplant (ASCT vs allo-HSCT): 62% vs 28%
• Treatment schedule:
  • IV nivolumab (3mg/kg) administered over 60 minutes every two weeks in the outpatient setting until progression, death, unacceptable toxicity, withdrawal of consent, or physician decision
  • Median exposure to nivolumab: 5.2 months (0–38.3)
  • Median cycles of nivolumab: 9.5
  • Discontinuations: 85.9% (progression [50.7%], consolidation with allo-HSCT [19.4%] and toxicity [9%])
• Primary endpoints: best response (ORR, CR, PR, stable disease [SD] and PD)
• Secondary endpoints: efficacy parameters such as duration of response, PFS and OS, as well as safety analysis
• Median follow-up: 34.3 months

Efficacy

The efficacy results for the whole cohort (n= 76) are shown in Table 2. The PFS rate was higher in patients achieving a CR with nivolumab compared to those achieving a PR, but no difference was seen in OS.

Table 2. Outcomes of whole cohort (n= 76)

Endpoint

Whole cohort

Confidence interval, range, p value

Best ORR

65.8%

90% CI, 55.82–74.80

CR

38.2%

-

PR

27.6%

-

Median time to response

2.6 months

95% CI, 2–4

Median time to best response

2.7 months

 

Median duration of response

24.3 months

95% CI, 9.9–not evaluable (NE)

Median PFS

12.1 months

95% CI, 7.3–26.2

3-year PFS

32%

-

PFS of patients in CR vs PR with nivolumab

Not reached vs 9.3 months

p< 0.001

Median OS

38.7 months

95% CI, 38.7–NE

3-year OS

65%

-

Table 3. Rates of relapse/progression by response to nivolumab

Response to nivolumab

Relapse or progression, N (%)

CR or PR (n= 50)

23 (46%)

CR (n= 29)

7 (24%)

PR (n= 21)

16 (76%)

Subsequent allo-HSCT consolidation

In total, 17 patients received allo-HSCT after nivolumab, 13 of whom had achieved an objective response (CR or PR) to nivolumab therapy. The four other patients transplanted had PD (n= 1) or had received salvage therapy (n= 3). Of the patients not in CR at transplant, five of six converted to CR post-allo-HSCT. At a median follow-up of 29.2 months, 13 patients of the 17 transplanted remained alive and disease-free. PFS and OS rates at one-year post-transplant were 76% and 82% respectively.

To evaluate if there was a benefit of consolidation with allo-HSCT, responses were compared between patients who responded to therapy and received transplant (n= 13) to those who responded but did not receive allo-HSCT (n= 37). At baseline, transplanted patients were younger and had a shorter duration of treatment with nivolumab, but there were no differences in relation to number of prior lines of therapy or disease stage. The time to best response was shorter in the transplanted group (2.4 months) compared to the non-transplanted group (1.8 months).

• Of 20 patients achieving a CR to nivolumab, 12 did not receive a subsequent allo-HSCT
  • Twelve patients (60%) remained disease-free at analysis
  • Conclusion: nivolumab monotherapy alone may be curative, even post-discontinuation
• However, 16 of 17 patients who did not achieve a CR with nivolumab monotherapy and did not undergo allo-HSCT eventually progressed
• Median PFS (transplanted vs not transplanted): NR vs4 months (95% CI, 7.3–32), p= 0.003
• Relapse rate (transplanted vs not transplanted): 0% versus2%, p< 0.001
• There was no difference in OS between the two groups.
• The PFS benefit may translate into an OS benefit with longer-term follow-up

Safety

Table 4 shows the AEs reported during the study. In total, 54 patients were alive at time of analysis. The main cause of death was lymphoma progression (n= 14, 58.3%)

Table 4. AEs reported during the study

AEs

N (%)

Total AEs

107

Number of patients experiencing an AE

40 (51.3%)

Grade ≥3 AEs

62

Number of patients experiencing a grade ≥3 AE

29 (37%)

Serious AEs (SAEs)

28

Number of patients experiencing a SAE

16 (20.5%)

Graft-versus­-host disease (GvHD)

• Of the 17 transplanted patients, all experienced GvHD
  • Acute GvHD: n= 14
    • Grade IV: n= 4
    • Steroid refractory: n= 5
  • Chronic GvHD: n= 7
    • Steroid refractory: n= 5

Treatment-related mortality (TRM)

• At day 100: 5.9%
• At 6 months: 11.8%
• At 12 months: 17.6%
• Deaths: n= 3

Conclusion

The study by Manson et al., has shown that in patients with R/R HL treated with nivolumab, most eventually progress despite high initial ORRs, especially those who do not achieve a CR. Patients who underwent allo-HSCT after nivolumab therapy however had a longer PFS compared to those who did not and had a lower relapse rate. At 34.3 months follow-up, this did not translate to a benefit in OS.¹ The results of this study show a benefit of allo-HSCT as consolidation following anti-PD1 therapy, however the results are interpreted cautiously due to the retrospective nature of the study and a limited patient sample. As there was no benefit in OS, it is essential to analyze the risk/benefit ratio and consider the possibility of re-sensitization with salvage chemotherapy as an alternative treatment option in this population. Prospective studies are required to identify the patients who benefit most from consolidation allo-HSCT after anti-PD1 therapy.

Meanwhile, the study by Armand et al., showed high PFS rates and cure rates can be achieved with pembrolizumab as post-ASCT consolidation in patients who do not respond to frontline therapy. The results also support the initiation of a further randomized trial of pembrolizumab as post-ASCT consolidation. Another study (NCT03057795) is investigating the combination of brentuximab vedotin with nivolumab as consolidation therapy for patients with R/R HL.