PET imaging: back in the game for gastric EMZL?

In this issue of Blood, Mayerhoefer et al¹ report the use of a novel PET radiotracer targeting the G-protein–coupled C-X-C chemokine receptor type 4 (CXCR4), for assessment of residual disease in gastric mucosa–associated lymphoid tissue (MALT) lymphoma after first-line Helicobacter pylori eradication.

Positron emission tomography-computed tomography (PET-CT), using ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) has generally been considered to be of little clinical utility for assessing extranodular marginal zone B-cell lymphoma (EMZL),² because of globally lower FDG avidity compared with aggressive lymphomas. Specifically, gastric primary MALT lymphoma has shown a low detection rate by FDG PET compared with other localizations, such as primary head-and-neck or bronchial lesions.³ This difference in detection rates was thought to be related to the thickness of the lesion. Thus, after eradication of H pylori, the follow-up is based on findings in pan upper gastrointestinal endoscopy, including multiple biopsy specimens obtained 3 months after treatment to rule out tumor progression with follow-up endoscopies (twice per year for 2 years, then every 12 to 18 months), to monitor the histological regression of the lymphoma.

CXCR4 and its ligand CXCL12 are overexpressed in a variety of tumor types and contribute to tumor growth, angiogenesis, metastasis, and therapeutic resistance. Recently, new imaging probes targeting CXCR4 have been developed for PET imaging, such as pentixafor labeled with gallium-68 (⁶⁸Ga-pentixafor), thereby enabling noninvasive imaging of CXCR4 expression throughout the whole body. In hematology, most experience with CXCR4-directed PET imaging has been in multiple myeloma⁴ and has shown the utility of this tracer. Recently, in patients with newly diagnosed MZL, Duell et al⁵ showed that CXCR4-directed imaging detects significantly more MZL localizations, particularly bone marrow infiltration, and affects the Ann Arbor classification, compared with conventional staging, including bone marrow biopsy, endoscopy, and FDG PET-CT.

The present innovative study further supports the use of CXCR4-directed PET imaging in MZL for response assessment. With time-matched gastric biopsy specimens used as the reference standard,^6,7 they found that ⁶⁸Ga-pentixafor PET accurately identified residual gastric disease in patients with MALT lymphoma with an impressive sensitivity of 95%. However, this high sensitivity is based on an observer-dependent visual assessment of the tracer uptake. Indeed, PET was rated as positive based on the visual increase in gastric uptake relative to the surrounding tissue. Although no false-positive cases were reported in this study, this criterion seems weak in an organ often associated with a slight diffuse physiological uptake and a relatively moderate pathological uptake. The lowest maximal standard unit value of biopsy-proven residual gastric MALT lymphomas in their cohort was 3.3. No blind interobserver reproducibility assessment was performed. Regarding the tumor-to-background ratios, the Deauville scale could be applied to increase the reproducibility in PET reporting between nuclear physicians. In addition, we wonder whether the different subtypes (ulcer, nodular, or gastritis; the subtype was not specified) or the presence of t(11;18)(p21;p21) translocation may have different PET appearances and whether they are associated with different sensitivity in the response. In summary, this study highlights the potential use of CXCR4 PET tracer for noninvasive response assessment in gastric lymphoma, although it warrants further studies to confirm this observation and to harmonize PET positivity criteria. In a broader context, this imaging may be more adaptable than FDG in some specific lymphoma subtypes, such as EMZL, because of the variable FDG avidity, or in central nervous system lymphoma,⁸ because of the intense brain physiological uptake. Further studies are warranted to better evaluate the utility of CXCR4-directed PET in diseases where we know that the response assessment criteria based on the Lugano criteria, widely used in aggressive lymphomas, are not perfectly adaptable.

Moreover, it has been shown that dysregulated expression of CXCR4 predicts disease progression in diffuse large B-cell lymphoma, independent of International Prognostic Index score and activated B cell/germinal center B-cell classification. In fact, CXCR4 overexpression impairs rituximab response, with an inverse correlation between the level of CXCR4 surface expression and the degree of rituximab sensitivity.⁹ In Burkitt lymphoma and chronic lymphocytic leukemia,¹⁰ the effect of rituximab was enhanced by antagonizing CXCR4. In this way, CXCR4-directed PET imaging should allow the visualization of CXCR4 overexpression through the whole body and thus may be complementary to FDG PET for risk stratification of patients with aggressive lymphoma.