PD-L1 Pathway Markers and Chromosome 9p24.1 Alterations in Patients with Classic Hodgkin Lymphoma Treated with Frontline Single Agent Pembrolizumab (PEM) Followed By AVD Chemotherapy

Background:

Chromosome 9p24.1/PD-L1/PD-L2 genomic copy number alterations (CNAs) including amplifications and copy number gains (CNGs) are predominant features of classic Hodgkin lymphoma (cHL) and lead to overexpression of the programmed death-1 (PD-1) ligands 1 and 2 (PD-L1 and PD-L2). Amplifications and high level CNGs have been associated with advanced stage cHL and inferior treatment outcomes with standard chemotherapy. There are few studies that correlate 9p24.1/PD-L1/PD-L2 amplifications or CNGs with response to PD-1 blockade monotherapy in the frontline setting. We conducted a phase 2 clinical trial of sequential pembrolizumab (PEM) x 3 followed by doxorubicin, vinblastine, dacarbazine (AVD) chemotherapy (4-6 cycles) for newly diagnosed cHL. Interim response to single agent PEM was assessed by PET-CT and by decline in metabolic tumor volume (MTV). Herein, we report the results of correlative studies analyzing 9p24.1 CNAs, PD-1 pathway expression and response to PD-1 blockade.

Methods:

Pre-treatment diagnostic biopsy specimens were double stained for PD-L1 (E1L3N, XP Cell Signaling) and PAX5, single stained for PD-L2 and pSTAT3, and scored by two expert hematopathologists (QC, LBS) for percentage positive cells and intensity of staining. A modified H score was calculated as the product of staining intensity (0-3) and percentage of positive tumor cells (0-100%), ranging from 0 - 300.

Fluorescence in situ hybridization to assess (FISH) chromosome 9p24.1 CNAs was performed by co-hybridizing PD-L1/PD-L2 probes (target) with the centromeric 9 probe (control). In each case, the percentage and magnitude of 9p24.1 CNAs were evaluated. Four FISH categories were defined based on the target: control ratio and the total copy numbers (CNs) of the target per Hodgkin Reed-Sternberg (HRS) cell to include: amplification (ratio≥3, CNs≥ 6), copy number gain (CNGs) (1≤ratio<3, 2<CNs <6), polysomy (ratio~1, CNs=3~5), and disomy (ratio=1, CNs=2). Patients were categorized according to the highest level of 9p24.1 alteration reaching at least 10% of counted cells. The relationships between PD-1 pathway markers, genomic alterations, and response to single agent PEM by MTV were assessed statistically using Fisher's Exact Test, Kruskal-Wallis test, and Spearman's Rank Correlation as appropriate. PD-L1 H Scores were grouped into terciles of approximately equal size for categorical analysis. Response was defined as a complete metabolic response (CMR), ≥ 90% reduction in MTV (near CMR), or partial response with < 90% reduction by MTV (PR).

Results:

Thirty patients were enrolled from September 2017, through August 1, 2019; 28 had tissue available for FISH analysis and 29 for immunohistochemistry. Response to single agent PEM was PR in 11 (36.7%), near-CMR in 8 (26.7%), and CMR in 11 (36.7%). CMR rate following AVD x 2 was 100%.

All patients in this analysis had genomic alterations, although 5 patients did not reach the cut off of 10%. The highest level alteration was amplification in 11 patients (36.7%), copy gain in 7 (23.3%), polysomy in 5 (16.7%) and disomy in 5 (16.7%) (Table 1). The average 9p24.1 copy number per HRS cell was 3.1 (range 2-8.1). Six of 22 examined cases were EBER-positive. There was no evidence of a statistical relationship between response to single agent PEM and 9p24.1 alteration, PD-L1, PD-L2 or STAT3 H-scores, or EBER status.

We found PD-L1 H Score tercile differed statistically by FISH category (P=.024, Fisher's Exact Test). Notably, there were no patients with amplification in the lowest tercile. More compelling, we found a positive association between PD-L1 H Score and average 9p24.1 locus copy number (Spearman's ρ=.36, P=.063, Fig 1a) and a negative association between PD-L1 H Score and percent disomic cells (Spearman's ρ= -0.21, P=0.29, Fig 1b) although the statistical significance of these results was limited by the small sample size.

Conclusions:

Our data is consistent with prior reports indicating a positive association between 9p24.1 genetic alterations and PD-L1 expression. The high response rates observed at all PD ligand levels seen in this clinical study suggests that even low levels of PD ligand expression may be sufficient for response to PD-1 blockade in previously untreated cHL.

View largeDownload slide

View largeDownload slide

Close modal