Pseudoprogression in Immunotherapy and Identification of Clinical-Biologic Predictive Factors: A Retrospective Pan-Tumor Cohort Study

Read the full article See related articles

Discuss this preprint

Start a discussion What are Sciety discussions?

Listed in

This article is not in any list yet, why not save it to one of your lists.
Log in to save this article

Abstract

Background Immune Checkpoint inhibitors (ICI) have emerged as one of the leading cancer therapies. Inflammatory mechanism induced by ICI can lead to increased tumor lesions that could be wrongly misinterpreted as progressive disease (PD). Currently, there are no imaging modalities that can differentiate true progression from pseudoprogression (PP). Although the iRECIST criteria have been developed to characterize these atypia, they appear to be inadequate in practice. The aim of this study is to determine whether this concept of PP exists and whether there are clinical or biological criteria that can be used to distinguish PP and true progression (PD). Methods We conducted a retrospective study in patients (pts) treated with ICI for metastatic cancer at Institut de Cancérologie de l’Ouest, and for whom PP was raised. Data were collected at initiation of ICI (t0), at evocation of PP (t1) and at subsequent evaluation (t2), treatment outcome and adverse events. Primary endpoint was to determine the percentage of pts with confirmed PP at t2. Secondary endpoints were to determine clinical or biological criteria associated to PP. Results 123 pts were included, 59% were men, most commonly with lung (38%), kidney (24%) or bladder (11%) cancer, and mainly treated in 2nd line with anti-PD1 (85%). We identified 56 confirmed PP (45%), with a median time of 79 days (28–227) between the first ICI infusion and evoked PP. ECOG score 0 and no weight loss were statistically associated with confirmed PP, respectively p < 0.0001 and p = 0.031. There was no association between PP and cancer type, metastatic site or type of ICI. For biological variable, only LDH was significantly different between the 2 groups (p = 0.003). More specifically, a 10% variation or more in LDH level was in favor of PD (OR = 0.72 with p = 0.005). Conclusions PP on ICI are real events and should be considered, regardless of tumor location, in pts in good general condition with stable LDH levels. Furthermore, with the development of artificial intelligence, some new imaging techniques could be developed to distinguish PP and PD. Some studies focusing on the circulating tumor DNA are also an interesting perspective.

Article activity feed