RANO 2.0 criteria: concepts applicable to the neuroradiologist’s clinical practice

Abstract

Purpose or review.

The Response Assessment in Neuro-Oncology (RANO) 2.0 criteria aim at improving the standardization and reliability of treatment response assessment in clinical trials studying CNS gliomas. This review presents the evidence supporting RANO 2.0 updates and discusses which concepts can be applicable to the clinical practice, particularly in the clinical radiographic reads.

Recent findings.

Updates in RANO 2.0 were supported by recent retrospective analyses of multicenter data from recent clinical trials. As proposed in RANO 2.0, in tumors receiving radiation therapy (RT), the post-RT MRI scan should be used as a reference baseline for the following scans, as opposed to the pre-RT scan, and radiographic findings suggesting progression within three months after RT completion should be verified with confirmatory scans. Volumetric assessments should be considered, when available, especially for low-grade gliomas, and the evaluation of non-enhancing disease should have a marginal role in glioblastoma. However, the radiographic reads in the clinical setting also benefit from aspects that lie outside RANO 2.0 criteria, such as qualitative evaluations, patient-specific clinical considerations, and advanced imaging.

Summary.

While RANO 2.0 criteria are meant for the standardization of the response assessment in clinical trials, some concepts have the potential to improve patients’ management in the clinical practice.

INTRODUCTION

Over the past decades, extensive efforts have been made to improve the standardization and accuracy of the longitudinal radiographic evaluations of gliomas for the detection of progression and treatment response [1–6]. Recently, the Response Assessment in Neuro-Oncology (RANO) 2.0 criteria have been proposed [6], which combine concepts from previous RANO high-grade gliomas (RANO-HGG) [3], RANO low-grade gliomas (RANO-LGG) [4], and modified RANO (mRANO) [5].

Of note, the RANO criteria are meant to improve the reliability and standardization of treatment response assessment in clinical trials. Therefore, although neuroradiologists are familiar with these criteria, they are not typically applied in the clinical practice. Indeed, the clinical radiographic reads also benefit from a less standardized approach, accounting for multiple nuanced aspects including qualitative evaluations and tailored considerations regarding the patient’s clinical status and treatment history. This review article discusses which concepts derived from the recent RANO 2.0 criteria can be applied to the clinical radiographic reads of serial follow-up imaging, under what circumstances, and with which caveats.

RECENT FINDINGS INFORMING RANO 2.0

Some of the novelties in RANO 2.0 were included after considering findings from recent studies. In a recent article, Youssef et al. [7] assessed whether some variations in the definition of progressive disease (PD) would result in a better correlation between progression-free survival (PFS) and overall survival (OS) in glioblastoma. Indeed, a certain definition of PD with better correlations with OS is arguably more adherent to a “true” biological progression. The correlation between PFS and OS was improved in newly-diagnosed glioblastoma when the post-radiation therapy (post-RT) scan was used as baseline, as opposed to the pre-RT scan. Additionally, the correlation was higher when confirmation scans for progression were used, and in particular the correlation was significantly improved in patients showing pseudoprogression (PsP) within 3 months from RT completion. Conversely, the evaluation of non-enhancing progression did not improve the correlation in this study, as already seen in a previous article by Huang et al. [8]. These important findings informed RANO 2.0, which now mandate the use of post-RT scans as baseline, require the use of confirmation scans for glioblastoma progression ≤3 months from RT completion, and waive the evaluation of non-enhancing progression in glioblastoma. Of note, confirmation scans for progression can be used even >3 months from RT completion, depending on the trial, for instance in case of trials employing immunotherapies. Similarly, the evaluation of non-enhancing progression in glioblastoma may be still desired in some cases, for instance for trials employing antiangiogenic treatments.

Another novel aspect in RANO 2.0 is the option of using volumetric measurements of the tumor burden with the threshold proposed in mRANO [5], as an alternative to bidimensional measurements. This option is supported by evidence in favor of volumetric measurements for the quantification of the tumor burden and/or the assessment of treatment response [9–13]. A recent study on slow-growing LGG by Ellingson et al. [11] suggested that monitoring tumor growth by means of volumetric assessments is preferable, as it increases the inter-reader agreement and reduces measurement inconsistencies across timepoints. An increased inter-reader agreement was also seen in studies comparing volumetric segmentations to manual diameters in glioblastoma, for instance by Blomstergren et al. [13].

RANO 2.0 CONCEPTS APPLICABLE TO CLINICAL RADIOGRAPHIC READS

RANO 2.0 provide rules to establish both PD and treatment response, which are useful to define PFS and objective response rates (ORR) in clinical trials, respectively. While PFS and ORR can both be informative as surrogate endpoints in clinical trials, in the clinical routine the crucial aspect for the patient’s management is the accurate identification of PD, rather than response. Indeed, in the absence of clinical or radiographic findings suspicious for PD during the longitudinal monitoring, the patient’s clinical management is typically left unchanged. In presence of radiographic progression, the current treatment regimen is usually interrupted, since it failed to control tumor growth and its continuation may cause toxicity in the absence of a potential clinical benefit. Similarly, in the case of young patients with LGG and a favorable molecular profile that may be kept under watch-and-wait strategies after a gross total resection [14], a radiographic progression mandates the suspension of this strategy and a switch to active treatment.

Therefore, clinical radiographic reads should be primarily aimed at correctly identifying PD, since a false positive classification of PD would prematurely interrupt a potentially effective treatment, while a false negative classification would maintain the patient on an ineffective regimen (with potential toxicity) and would possibly prevent a switch to a more effective therapy. Thus, this article focuses on RANO 2.0 rules and concepts that concern the definitions of progression (Table 1), rather than response.

Table 1.

Summarized overview of RANO 2.0 to define progression in clinical trials.

Baseline scan for the longitudinal monitoring
Newly diagnosed tumors
Receiving RT (e.g., glioblastoma)	Post-RT scan
Not receiving RT (e.g., non-CE LGG)	Pre-Tx scan, not post-surgical
Recurrent tumors	Pre-Tx scan, ideally “hyperacute” baseline*
Reference scan for comparison to define PD
Without confirmation scans	Compare to baseline scan or nadir
With confirmation scans
Preliminary PD	Compare to baseline scan or nadir
Confirmation of PD	Compare to preliminary PD, PsP, or nadir-after-PsP°
Tissue under evaluation (CE vs non-CE)
CE tumors	Evaluate only CE tissue, (except if receiving antiangiogenic Tx, possibly)
Non-CE tumors	Evaluate non-CE tissue and the appearance of CE
“Mixed” tumors	Evaluate non-CE tissue and CE-tissue
Measurement technique
Sum of bidimensional products of all target lesions or volumetric segmentations of the tumor
Definition of radiographic progressive disease (PD) or preliminary PD
In the absence of measurable disease:
Appearance of new measurable disease (if glioblastoma, only new CE disease) or significant growth of previously non-measurable disease becoming measurable:
In presence of measurable disease:
≥25% increase in the sum of bidimensional products (or ≥40% in volume) of target lesions
When to use confirmation scans for PD (performed after ≥4 weeks)
Required	Within ≤12 weeks after RT completion
Can be considered	In recurrent tumors with PD at first follow-up*
Optional / not required	For non-CE PD in non-CE tumors
Influence of clinical/neurological status on PD assessment
Clinical/neurological deterioration alone determines PD, regardless of MRI findings, unless attributable to causes other than the tumor
Influence of corticosteroids on PD assessment
Corticosteroid dose decrease can justify a downgrading of radiographic PD to SD
Corticosteroid dose increase alone does not qualify as PD

^*In the recurrent setting, tumor growth may occur between the pre-Tx scan and the Tx start. Ideally, a “hyperacute” pre-Tx scan may be obtained immediately before Tx. If not feasible, requesting confirmation scans in case of PD at the first follow-up may be considered.

^°Once a preliminary PD is seen, PD can be confirmed at the following timepoint (≥4 weeks), if additional tumor growth is seen, in which case confirmed PD is backdated to the date of preliminary PD. Alternatively, if additional tumor growth is not seen at the following timepoint, preliminary PD is labeled as PsP. Then, PD is confirmed without backdating if any following timepoint shows tumor growth compared to either PsP or to the nadir-after-PsP (if present).

CE = contrast-enhancing, PD = progressive disease, PsP = pseudoprogression, RT = radiation therapy, Tx = treatment.

Baseline scan for Longitudinal Monitoring

RANO 2.0 mandate the use of post-RT scan as baseline for the longitudinal monitoring of newly-diagnosed gliomas receiving RT. This minimizes the confounding effects arising from radiographic changes happening in the post-RT scan as a result of chemoradiation [5]. In clinical practice, neuroradiologists and clinicians should take this into consideration and use the post-RT scan as reference baseline to compare further scans, while refraining from classifying as PD any worsening of the radiographic findings observed between pre- and post-RT scans.

For patients not undergoing RT, RANO 2.0 propose to not use as baseline the early post-surgical scan (<72h) that is typically used to assess the extent of resection and post-surgical complications. Instead, it is preferable to use a later scan as baseline, before the initiation of any treatments (pre-Tx), to allow for the resolution/stabilization of the post-surgical changes before starting the radiographic monitoring. Post-surgical changes can mimic PD, as they can present as new areas of T₂/FLAIR-hyperintensity due to post-surgical edema or gliosis [15], or even as new areas of non-tumoral contrast-enhancement near the cavity wall due to blood-brain barrier permeability, possibly linked to reactive changes and post-surgical neovascularization [16, 17]. This should be considered also in the clinical practice. Pragmatically, since tumors not receiving RT are usually slow-growing lesions with a favorable molecular profile that are scanned at ~3 months at the beginning of follow-up [14], the 3-month scan after surgery may be used as baseline for the following longitudinal monitoring, as long as no treatment is initiated beforehand. In this case, potential radiographic changes seen between the post-surgical scan and the 3-month follow-up should not be evaluated as PD.

In the recurrent setting, the readers should be aware that there may be interval tumor growth between the scan where the recurrence is noted and the start of the new treatment, resulting in worsened radiographic findings on the first follow-up scan. RANO 2.0 propose to account for this instance by either obtaining “hyperacute” baseline scans 1–2 days before the start of the new treatment or, if that is not feasible, obtaining a baseline MRI ≤2 weeks prior to initiation of therapy and consider using confirmation scans in case progression is seen on the first follow-up. Either one of these measures may be applied to the clinical practice, when possible.

Tissue under evaluation: Enhancing and/or Non-Enhancing Tissue

RANO 2.0 proposes different sets of rules depending on whether a glioma: 1) is predominantly enhancing, 2) is non-enhancing, and 3) has both enhancing and non-enhancing components (“mixed” gliomas). In clinical trials, applying the “predominantly enhancing” or “mixed” criteria depends on the sponsor, based on the molecular types and grades included in a certain trial. In the clinical practice, the reader can decide on a patient-specific basis, as long as the same criteria are used consistently across all timepoints. Overall, the “mixed” criteria should be reserved for lesions showing predominant non-enhancing neoplastic tissue that only presents some spots or nodules of contrast-enhancement, while all the other lesions with larger enhancing components may be evaluated as “predominantly enhancing”.

In non-enhancing gliomas, non-enhancing tissue should be evaluated to assess the presence/absence of PD. Additionally, any new measurable enhancing component in a previously non-enhancing lesion should be considered PD. In “mixed” gliomas, both non-enhancing tissue and enhancing tissue should be evaluated separately, and PD is assigned if either a non-enhancing or enhancing progression is seen. In predominantly enhancing gliomas, the progression of the contrast-enhancing components is undoubtedly the most important aspect to value. RANO 2.0 propose to not evaluate the non-enhancing progression of enhancing tumors, except optionally in lesions treated with medications that significantly alter the blood-brain barrier permeability (e.g., antiangiogenic agents). This choice is justified for clinical trials, since it improves the standardization of the criteria by eliminating the subjectivity inherently linked to the evaluation of non-enhancing progression, which is hard to measure and confounded by the presence of vasogenic edema and radiation-induced signal alterations [18]. While clinical radiographic reads should adopt from RANO 2.0 the general concept that PD in enhancing gliomas should be primarily dependent on the growth of enhancing tissue, it is reasonable to leave room for the identification of unequivocal non-enhancing progression in the clinical setting. Unequivocal non-enhancing progression may be represented by a new area of T₂/FLAIR signal alteration located in previously normal-appearing grey matter (e.g., cortex or thalamus) and associated with signs of mass effect (e.g., gyral enlargement, sulcal effacement) (Figure 1A). Additional qualitative considerations can strengthen the hypothesis of non-enhancing progression, such as the presence of tissue with progressive diffusion restriction not ascribable to intralesional hemorrhage (ruled out with susceptibility- or T₂*-weighted images), which further points to solid tissue growth.

Figure 1. Representative cases to discuss the adoption of RANO 2.0 concepts in clinical radiographic reads for glioblastoma.

While RANO 2.0 recommend to abandon the evaluation of non-enhancing progression of glioblastoma in clinical trials, it is reasonable that unequivocal non-enhancing progression is still reported in the clinical routine (e.g., a new cortical T₂-FLAIR signal alteration, arrowhead in A). Of note, subsequent follow-up scans showed enhancing progression in the same region (arrows in A). RANO-defined pseudoprogression consists in growth or new appearance of CE tissue that does not additional grow on the confirmation scan obtained after ≥4 weeks (arrows in B). Conversely, PD is confirmed when the confirmation scan shows additional growth compared to preliminary PD (arrows in C).

CE = contrast-enhancing, PD = progressive disease, PsP = pseudoprogression.

While too subjective for standardized criteria in clinical trials, these examples of non-enhancing progression may be applicable to the clinical routine, where a higher degree of subjective and qualitative considerations is allowed. In any case, T₂/FLAIR signal alterations should not be tracked with measurements in predominantly enhancing gliomas, as the assessments would be largely influenced by the aforementioned non-neoplastic alterations. Importantly, neuroradiologists should be aware that clinical trials enrolling patients with recurrent glioblastoma usually require the evidence of enhancing progression as an inclusion criterion. Therefore, categorizing a certain follow-up scan as PD based on T₂/FLAIR findings may warrant re-treatment while likely precluding the patient’s access to recurrent clinical trials (unless treatment is withheld until enhancing progression is seen).

Definition of Radiographic Progressive Disease

In RANO 2.0, “measurable disease” is defined as a lesion with clear boundaries and a size of ≥1x≥1x≥1 cm on 3D imaging, or ≥1x≥1 cm in-plane on two consecutive slices on 2D imaging – excluding surgical cavity, necrotic areas, and cysts. If a lesion is measurable, it can be used for the quantification of tumor burden, which is calculated with segmentations (3D method) or as the sum of the products of bidimensional in-plane diameters (2D method). As mentioned, for enhancing tumors, only enhancing tissue is evaluated for the assessment of measurable disease and tumor burden.

If measurable disease is present at baseline and nadir, then a ≥25% (2D) or ≥40% (3D) increase of the tumor burden is considered radiographic evidence of progression, which can be labeled either as PD (when confirmation scans are not required) or preliminary PD (when confirmation scans are required). If no measurable disease is present at baseline or at nadir, PD/preliminary PD is assigned when a follow-up scan shows measurable disease. When confirmation scans are required, an additional ≥25% (2D) or ≥40% (3D) increase of the tumor burden in a confirmation scan (≥4 weeks) compared to preliminary PD is required to re-classify preliminary PD as confirmed PD. If this condition is not met, preliminary PD is re-classified as confirmed PsP, and confirmed PD will only be assigned in case of further radiographic evidence of progression in subsequent scans compared to the nadir-after-PsP or to PsP, without additional confirmation scans. Importantly, this concept is substantially similar (with some differences) to what is proposed in the BT-RADS system (btrads.com/resources) for the scoring of the post-treatment radiographic findings in gliomas [19–21], where the category “highly suspicious for PD” (BT-4) is reserved to cases showing progressively worsened radiographic findings over two subsequent scans.

This RANO 2.0 flow-chart for PD may be too rigid for the clinical practice, but can still be informative. In clinical radiographic reads there is no need to strictly apply the ≥25% (2D) or ≥40% (3D) threshold to define PD, and a more subtle size increase could also be considered PD. However, these thresholds should inform on the order of magnitude of changes defining PD, i.e. a size increase of a few millimeters should still be considered as stable disease (SD) in most cases. Similarly, readers may not strictly apply the definition of “measurable disease” as ≥1x≥1x≥1 cm, but they should be aware that measures of millimetric lesions should not be tracked, mainly because technical constraints determine oscillations in the measurements of small lesions, with the risk of overcalling PD.

The RANO 2.0 flow-chart also acts as a reminder of which scan should be used as reference to assess whether the subsequent scans display progressive findings. At the start of the follow-up, the reference scan for comparison is always the baseline scan. If the tumor burden shows any size decrease over time, the reference scan to assess progression becomes the nadir. In case radiographic progression is seen and confirmation scans are required, the scan with preliminary PD becomes the reference scan for the subsequent confirmation scan, meaning that PD is confirmed only if additional growth is seen compared to preliminary PD (Figure 1B–C). This is important because true PD is expected to continue growing in time (Figure 1B), while PsP either shrinks or stabilizes (Figure 1C). In case PsP is confirmed, the PsP scan becomes the new reference scan for the future definition of progression. Finally, if the tumor burden shows a size decrease after PsP, the following scans should be compared to the nadir-after-PsP. While being aware of the single RANO 2.0 reference scan is important, radiographic clinical reads also benefit from a higher-order evaluation of trends seen across multiple scans. For instance, in the clinical setting PD may be called based on an overall growth tendency even though a RANO 2.0 rule for PD is not technically met (Figure 2A). Similarly, unlike in clinical trials, progression can be called in a clinical setting also based on the qualitative observation of growing solid tissue causing mass effect or unequivocal infiltrative growth, without a strict need to demonstrate lesion increase using RANO-style measurements, especially in lesions that are difficult to measure (Figure 2B).

Figure 2. Representative cases to discuss the adoption of RANO 2.0 concepts in clinical radiographic reads for low-grade gliomas (LGG).

When the baseline scan shows no measurable disease, radiographic progression is defined by the appearance of RANO-defined measurable disease (≥1 × ≥1 cm with in-plane perpendicular diameters, on two consecutive slices) (A). While this rule should be applied in clinical trials, in the clinical routine a more subjective approach that takes into account the gradual changes over time may also be adopted (A). When the baseline scan shows the presence of measurable disease, the tumor burden should be tracked with quantitative measurements in clinical trials, in order to evaluate to evaluate whether the lesion size change meets criteria for radiographic progression. In the clinical practice, a more qualitative and subjective approach may also be adopted, for instance when assessing lesions with infiltrative growth pattern, that are challenging to measure (B). Qualitative hallmarks of tumor growth include progressive thickening of grey matter structures (arrow in B), infiltrative growth causing an overall tissue enlargement (asterisks in B), and the effacement of CSF spaces due to mass effect (arrowheads in C).

Another scenario where the clinical reads may adopt concepts from RANO 2.0 is the appearance of new contrast-enhancement during the follow-up of non-enhancing gliomas. In such case, it may be reasonable to wait for the contrast-enhancing area to become measurable before calling PD. A feasible strategy in the case of non-measurable new enhancement may be to require a follow-up scan at 1 month to assess whether the enhancement grows and becomes measurable.

When to use Confirmation Scans for Progressive Disease

The rationale behind confirmation scans in RANO 2.0 is to mitigate the effects of PsP on false positive classifications of PD in case of suspicious enhancing progression. RANO-defined PsP can be broadly defined as any worsened radiographic finding that does not keep worsening over time, but rather stabilizes or resolves, which definition comprises several biological and histological entities [22, 23]. Based on the scenarios when PsP can be more frequent, confirmation scans in RANO 2.0 are recommended for confirmation of enhancing progression ≤3 months after RT completion or during treatments with specific agents (e.g., immunotherapies). In the clinical practice, neuroradiologists should consider these scenarios and consider waiting for confirmation scans before formally calling progression, and possibly order a confirmation scan for an earlier date than originally scheduled. Of note, this approach is also similar to what is proposed by BT-RADS in this scenario, according to which worsened radiographic findings ≤3 months after RT can be interpreted as possible treatment effects (“favor treatment”, BT-3a), and advocate for a decreased time interval of follow-up without treatment change. In general, as the clinical routine allows more flexibility, neuroradiologists may consider confirmation scans whenever encountering non-univocal findings. When using confirmation scans, two rules from RANO 2.0 should always be observed. First, the confirmation scans must be obtained ≥4 weeks after the preliminary PD findings. Second, the scan with preliminary PD should be used as reference to determine whether the confirmation scan shows additional progression (confirmed PD) or not (confirmed PsP). Conversely, in presence of non-enhancing progression in non-enhancing gliomas (e.g., LGG), confirmation scans have no clear usefulness, as this scenario is unlikely to represent a PsP, therefore PD can be defined on the first scan showing significant tumor growth.

Neurological/Clinical Status and Corticosteroid Dose

In RANO 2.0, a clear neurological worsening, assessed with scores such as NANO, or a clear deterioration in the overall clinical status, evaluated with Karnofsky Performance Status or ECOG, are sufficient to determine PD, regardless of the radiographic findings. A clinical radiographic read should only focus on the interpretation of the images, therefore a radiology report should not call progression depending on clinical findings only. However, it is reasonable to use neurological/clinical status to improve the interpretation of radiographic findings that hint towards progression.

As for corticosteroids, in RANO 2.0 the only scenario in which a change in corticosteroid dose impacts the definition of progression is when radiographic findings are worsened and corticosteroids are decreased compared to the reference scan. In this case, PD is downgraded to SD because the change in steroid dose may explain the observed radiographic worsening. In the clinical setting, neuroradiologists should take this concept into consideration and make sure to collect information on corticosteroid dose at the time of the scan, in order to better interpret the findings. While RANO 2.0 suggests a standardized categorical approach for clinical trials, where corticosteroid dose is noted as “stable” vs “decreased” vs “increased”, the clinical radiographic reads may benefit from a more nuanced approach, where the neuroradiologist may evaluate whether the degree of radiographic worsening may be entirely justified by the magnitude of the corticosteroid dose change. However, such an evaluation is rather arbitrary and should be conducted cautiously, since literature quantifying tumor size changes following corticosteroid dose change is still very sparse [24], and mainly focusing on changes in diffusion [25] and vasogenic edema [26].

ADVANCED IMAGING

RANO 2.0 criteria do not include the evaluation of advanced imaging, mainly due to a lack of multicenter standardization and the current absence of reliable universal thresholds for their interpretation. However, advanced MR techniques are rather frequently employed in the longitudinal monitoring of gliomas, especially to differentiate true progression from pseudoprogression, as recently reviewed [27, 28]. For this application, perfusion studies are probably the most widely employed, and their usefulness is supported by several meta-analyses [29–32]. An elevated relative cerebral blood volume (rCBV) is generally considered the most robust advanced biomarker for true progression in the clinical setting, reflecting tumoral angiogenesis. Additionally, amino acid PET (e.g., DOPA-PET and FET-PET) can also be useful in this scenario [33, 34]. Although not currently included in treatment response criteria for clinical trials, in the clinical practice each center should consider using one or more advanced techniques to support the longitudinal monitoring of gliomas, depending on the institutional availability and expertise.

PRACTICAL GUIDE FOR RANO 2.0 APPLICATION

While the formal RANO 2.0 article lists all RANO 2.0 criteria in detail [6], the reader may also refer to a recent article providing step-by-step guidance on how to operationalize RANO 2.0 criteria [35].

CONCLUSIONS

RANO 2.0 criteria are meant for the standardization of radiographic reads for clinical trials, rather than for the application in the clinical practice. However, neuroradiologists performing clinical radiographic evaluations should be aware of RANO 2.0, as they can be informative for the selection of the baseline and reference scan, for the decisions on when to order confirmation scans and for providing insights on how to interpret them, and for the consideration of clinical/neurological status and corticosteroid doses when evaluating radiographic findings. Still, the clinical practice also benefits from complementary qualitative and nuanced evaluations of brain scans, which are not included in standardized response assessments.

Key Points:

• Informed by recent studies and by expert consensus, RANO 2.0 criteria are aimed at standardizing the monitoring of gliomas receiving treatments

• Neuroradiologists should be aware of RANO 2.0 criteria, and especially of the RANO 2.0 rules that define disease progression, which warrants a change of treatment in the clinical practice

• The definition of disease progression according to RANO 2.0 should take into consideration the timing from radiation therapy, should consider the use of confirmation scans, and should account for corticosteroid dose and clinical status