In a prior article reviewing cancer immunotherapy innovations we presented the successful practice surgeon William Coley applied to the intratumoral injection of “Coley’s toxins” into sarcomas over 100 years ago, which has continued in medical practice to this day.1,2 Traditional oncology response criteria, i.e., WHO or RECIST, are based on assessing tumor response to systemic chemotherapies, and over the years there have been adaptive modifications of the criteria such as iRECIST, mRECIST and irRECIST. Notably, Dr. Gregory Goldmacher recently introduced new guidelines to delineate potential response differences between injected and noninjected tumors, referred to as itRECIST (Intratumoral Immunotherapy Response Criteria in Solid Tumors) during the American Society of Clinical Oncology’s annual meeting (ASCO 2020).3
The proposed response criteria are tailored for solid tumor and intratumoral therapy trials and incorporate features of RECIST and iRECIST with the goal of explaining the distinct effects expected locally and systemically in these trials. As with RECIST the classification algorithm follows the measurability (quantitative) and non-measurability (qualitative) assessment of tumor burden and defines the maximal number of target lesion (TL) tumors using the same measuring criteria as within RECIST 1.1. Importantly, the new criteria further categorize the target and nontarget tumor categories into injected and noninjected tumors (i.e., target injected (T-I), target noninjected (T-NI), nontarget injected (NT-I), and nontarget noninjected (NT-NI)) and defines follow up of t 10 TL (i.e., up to 5 T-I and up to 5 T-NI) with all other lesions defined at NTL (i.e., NT-I or NT-NI).
The paper provides standardized recommendations to aid clinicians in the clinical treatment planning regarding the prioritization and selection of lesions best suited for injection. The injected lesions, T-I and NT-I at baseline, will “each have their own distinct sum of diameters (SOD; longest diameter is used for measuring extranodal lesions, and the short axis is measured for lymph nodes as defined in RECIST 1.1). A combined SOD also includes all target lesions – injected and noninjected. Nontarget injected (NT-I) and nontarget noninjected (NT-NI) are observed qualitatively, exactly as in RECIST 1.1.”3 The combined responses of these lesions, new lesions and the reclassification of any noninjected tumors after baseline assessments, which were subsequently injected, have been accommodated for in the modified response classification tables. (Readers are referred to Dr Goldmacher’s paper for further details.3) It’s important to note that T-I lesions may change in size at each treatment visit and change assessments are made by comparing to the last image evaluation time point. Additionally, noninjected target lesions follow standard RECIST rules comparing the decrease in SOD from baseline and increase in SOD in relation to nadir for Partial Response (PR) and Progressive Disease (PD), respectively.
Dr Goldmacher’s paper suggests following iRECIST rules when deciding to continue immunotherapy at the time of PD diagnosis when patients are stable. It also provides an algorithm for treatment and continued injection when variable response patterns (pseudoprogession or appearance of new lesions) are observed.
As the scientific and regulatory community begins to debate the merits of these newly introduced criteria, imaging core labs will work to operationalize the itRECIST response assessments into standardized oncology assessment workflows and platforms. This will serve to streamline readers’ workloads and facilitate lesion tracking of potential response differences between injected and noninjected lesions. The result will be to allow independent readers to perform multiple response criteria assessments simultaneously to compare the best predictor of outcome in novel intratumoral treatment paradigms.
- Am. J. Med. Sci. 131: 375–430
- J Immunol 2017; 198:31-39
- J Clinical Oncology Volume 38, Issue 23 2667