Concurrent Monitoring of Peripheral Blood Circulating Tumor DNA and Circulating Tumor Cells in Relapsed/Refractory Follicular Lymphoma Patients Post Axicabtagene Ciloleucel, a Single Center Experience

Background:

Monitoring of peripheral blood (PB) circulating tumor DNA (ctDNA) in DLBCL post axi-cel can risk stratify and predict outcomes in DLBCL (Frank et al, JCO 2021). The utility of monitoring minimal residual disease (MRD) in follicular lymphoma (FL) post-CART therapy is currently unknown and the correlation between ctDNA and circulating tumor cells (CTC) as a MRD tool in this context is not yet determined. We present data from a retrospective pilot study to detect MRD by ctDNA (from banked serum) and CTC (from banked PB mononuclear cells, PBMCs) post axi-cel therapy in patients (pts) with R/R FL pts enrolled on the Zuma-5 trial at Moffitt Cancer Center.

Methods:

Lymphoma clonotype(s) were identified in extracted tumor DNA from banked archival FFPE tissue by NGS (clonoSEQ®;Adaptive Biotechnologies) to identify rearranged IgH (VDJ), IgH (DJ), IgK and IgL receptor gene sequences. MRD was monitored with PCR and NGS (clonoSEQ) in banked serum (ctDNA) and PBMCs (CTC) samples (pre-lymphodepletion, weekly during the first month, and subsequently every 3 months (mo) up to 2 years when available). We compared MRD measured by ctDNA versus CTC and correlated both with clinical responses by PET imaging per Lugano criteria (performed at day 30, then every 3 months post axi-cel infusion).

Results

Fifteen pts received axi-cel for R/R FL between June 2018 and July 2019. One pt failed to have a clone suitable for MRD tracking and was excluded from the current analysis. After a median follow up of 31.7 mo, the median PFS and OS were not reached and the estimated PFS and OS at 2.5 years were 57% and 85%, respectively. Six pts developed disease progression, all but one in the first year with that patient progressing at 18 mo. PET response at day 30 was CR in 8 pts (57%) and PR in 6 pts (43%). Six pts developed PD at a median of 7 mo post axi-cel infusion (range 3-19 mo) and 5 of them received a second infusion of axi-cel (3 had durable remissions afterwards). Prior to CART infusion, MRD was detected in 55% (cfDNA) and 64% (CTC) of available baseline samples. MRD was cleared rapidly within the first 2 weeks post axi-cel infusion, and MRD positive rates were as follows: day 7: ctDNA=54%, CTC 29%; day 14: ctDNA 17%, CTC 7%. Thus, ctDNA was more frequently positive than CTC when done at the same time points. All available samples at day 14, day 21, and day 28 were MRD negative when assessed by ctDNA or CTC. There was no difference in the incidence of disease relapse, response by PET at day 30, or PFS between MRD negative and MRD positive pts at days 7 or 14 when assessed by ctDNA or CTC. There was no correlation between the MRD clone copy numbers prior to lymphodepletion and responses assessed by PET, PFS, severity of cytokine release, or severity of neurotoxicity. Ten pts had available samples beyond day 28 post axi-cel infusion (median 192 days, range 90-418 days), of which 5/10 later developed PD including 2 pts (see figure 1) with banked samples close to the time of relapse. In both of these 2 pts, MRD was detectable at the time of progression on PET (n=1) or 3 mo prior to progression on PET (detected by ctDNA but not CTC). Both of those pts were retreated with axi-cel upon evidence of PD on PET. Otherwise pts who remained in CR continued to have negative MRD by ctDNA and CTC (n=5). There was only one patient who developed PD without detectable MRD at baseline or the time of relapse.

Conclusion

To our knowledge, this is the first study to assess MRD measured by ctDNA and CTC at concomitant time points in pts with FL post CART therapy. Our pilot study suggests that ctDNA is more sensitive than CTC to detect MRD in PB. Unlike reported experiences in DLBCL post-CART, in our small cohort, detectable MRD at day 7 did not predict future relapse or correlate with PFS. However, MRD by ctDNA was detected in pts prior to clinical or radiologic progression of disease. A prospective ctDNA MRD monitoring study in FL is warranted and an MRD driven early retreatment CART trial should be considered.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal