Therapeutic response assessment of high-grade gliomas during early phase drug development in the era of molecular and immunotherapies

Abstract

Several new therapeutic strategies have emerged over the past decades to address unmet clinical needs in high grade gliomas, including targeted molecular agents and various forms of immunotherapy. Each of these strategies requires addressing fundamental questions depending on the stage of drug development, including ensuring drug penetration into the brain, engagement of the drug with the desired target, biologic effects downstream from the target including metabolic and/or physiologic changes, and identifying evidence of clinical activity that could be expanded upon to increase the likelihood of a meaningful survival benefit. The current review article highlights these strategies and outlines how imaging technology can be used for therapeutic response evaluation in both targeted and immunotherapies in early phases of drug development in high grade gliomas.

Current Therapeutic Landscape for High Grade Gliomas: Molecular and Immunotherapies

Several new therapeutic strategies have emerged over the past 20 years to address unmet clinical needs, particularly in recurrent disease in which there is no consensus as to the standard of care as no therapeutic options that have produced substantial survival benefit¹, including the development of specific targeted agents and various forms of immunotherapy. Each of these therapeutic strategies and stages of clinical trial evaluation have their own unique questions, potential complications, and specific mechanisms of action that can be distinctively addressed using imaging techniques and approaches. For example, early phase trials using targeted agents might benefit from having information about blood-brain barrier permeability, drug penetration into the brain, or downstream physiological changes known to accompany the treatment, whereas later stage trials might be more focused on tumor shrinkage, growth rate changes using serial measures of bulk enhancing or non-enhancing tumor, and overall survival (Fig. 1). During immunotherapy development investigators may have unique challenges, including differentiation of treatment-related inflammation (i.e. “pseudoprogression”) from progressive disease and spatial differentiation and/or quantification of particular immune cells (e.g. CD8+ or macrophages with a particular polarity) from active tumor cells within a heterogeneous enhancing mass. The current review article will highlight some of these strategies and outline how imaging technology can be used for therapeutic response evaluation in both targeted and immunotherapies.

Fig. 1. Relevant Questions and Imaging Strategies for Stages of Drug Development.

Each stage of drug development for high grade gliomas (Phase 0 through Phage 3) has unique questions as well as ways imaging technology can be used to address these questions. During Phase 0 studies questions regarding blood brain barrier (BBB) penetration, target engagement, and downstream biologic effects of the drug are most relevant, which can be quantified through use of radiolabeled drug, metabolic and physiologic imaging technologies. During phase 1 studies, evidence of biologic effects that increase likelihood of clinical activity and evidence of direct clinical activity are pertinent, which can be identified through sustained metabolic or physiologic imaging changes, evidence of inhibited macroscopic (bulk tumor) or microscopic (proliferation) growth rate. In later stage (phase 2-3) trials, clinical activity and increases likelihood of a meaningful clinical benefit and direct evidence of clinical benefit are important, which can be recognized through use of landmark progression-free survival (PFS) benchmarks, sustained growth rate inhibition, and increased overall survival.

Identifying BBB Penetration and Target Engagement in Molecular and Immunotherapies

Drug delivery, brain penetration, and target engagement are significant limitations in neuro-oncology drug development for both targeted therapies² and immunotherapies³, due in part to the blood-brain barrier (BBB)⁴ and other factors. Depending on the molecular, chemical, and physical characteristics of the therapeutic agent under investigation, it may require transport of the agent via paracellular or transcellular diffusion, carrier- or adsorptive-mediated transport, or receptor-mediated transport⁵. The size of the molecule, charge distribution, and other factors must all be considered when developing agents specific for neuro-oncology⁶, which may explain the large rate of failure of most agents that have been shown to be effective and were designed for other systemic solid tumors⁷.

Evaluation of BBB permeability and initial target engagement for small molecules can be evaluated several ways using advanced imaging techniques early in the drug development process. Dynamic contrast enhanced (DCE) perfusion MRI, a technique utilizing serial T1-weighted MRI during injection of gadolinium-based contrast agents, is often used to quantify exchange of contrast agents between the vascular and extravascular space in a range of diseases⁸. The exchange rate of contrast moving from the vascular to extravascular space, or K^trans, has been shown to be dependent on capillary permeability and surface area^8,9 and can be used as a surrogate of BBB permeability for the purposes of drug development and understanding drug penetration¹⁰. However, the BBB is highly selective, therefore this approach may not reflect BBB permeability to the investigative therapeutic agent, but rather only highlights permeability to the gadolinium-based contrast agent used. An alternative approach may be to radiolabel the investigative agent with a single photon or positron emitting nuclei (e.g. SPECT nuclei ^99mTc or PET nuclei like ¹¹C or ¹⁸F), then evaluate the distribution after infusing this tracer into the patient. For example, Gerstner et al.¹¹ investigated drug penetration of temozolomide, an alkylating agent shown to have significant clinical activity in GBM, using radiolabeled [¹¹C]-temozolomide, DCE-MRI, and dynamic susceptibility contrast (DSC) perfusion MRI in recurrent GBM patients treated concurrently with bevacizumab, an anti-angiogenic agent known to reduce BBB permeability. Results from this study showed that in areas where BBB permeability was not compromised on DCE-MRI, increased blood flow on DSC-MRI was sufficient to increase temozolomide uptake within brain tumors. A recent study by Jucaite et al.¹² is another example of this approach being used to successfully determine brain exposure to an experimental therapy. In this study, investigators used microdoses of ¹¹C-AZD1390, a radiolabeled version of a potent and selective ATM inhibitor AZD1390, to measure brain exposure in healthy volunteers and noted the agent crosses the BBB noting that the agent crosses the BBB and has favorable temporal kinetics throughout the brain. This information can be useful for guiding dose ranges and schedules for subsequent clinical studies in order to optimize exposure. This same strategy can be used with most targeted and immunotherapy agents, as many undergo similar early bioavailability, biodistribution, and safety studies using radiolabeled versions of the compound of interest^13,14, including peptides and antibodies¹⁵. Radiolabeled versions of common chemotherapies, targeted agents and immunotherapy agents in use today include temozolomide¹¹ (e.g. ¹¹C-temozolomide), lomustine¹⁶ (e.g. ¹⁴C-lomustine), bevacizumab¹⁷ (e.g. ⁸⁹Zr-bevacizumab, ¹¹¹In-bevacziumab, or ¹²⁴I-bevacizumab), cediranib¹⁸ (e.g. ¹⁴C-cediranib), nivolumab^19,20 (e.g. ⁸⁹Zr-nivolumab, ⁶⁸Ga-nivolumab), and ipilimumab^21,22 (e.g. ⁶⁴Cu-ipilimumab).

Radiolabeled drugs can provide important information regarding drug penetration and delivery to brain tumors, it does not necessarily address target engagement (Fig. 1). While there are a number of established methods for quantifying target engagement ex vivo and in vivo within living cells or systems^23-26 and temporal characteristics of radiolabeled drug uptake and distribution can be used to infer target engagement using non-labeled drug at increasing concentration to infer receptor characteristics (e.g. ^27,28), imaging of direct target engagement remains a significant challenge in human patients²⁹. While still a proof-of-concept, ³¹P-NMR spectroscopic approaches could provide value in understanding phosphorylation of various receptor proteins through the NMR chemical shift of phosphorus nuclei^30,31, but this approach has not been shown to be sensitive enough in human brain tumor drug development to date.

Several molecular imaging tracers have also been developed to quantify the presence or activation of T-cell surface markers for immunotherapies in both murine models and humans (reviewed in ³²), including both antibody fragments and cytokines. For example, presence of T-cell receptors (TCR) CD4 and CD8 have been explored using portions of antibodies, or cys-diabodies^9,33 (e.g. ⁸⁹Zr-malDFO-GK1.5 cDb and ⁸⁹Zr-malDFO-169 cDb, respectively) and minibodies^34-36 (e.g. ⁸⁹Zr-Df0IAB22M2C). Additionally, strategies to identify target engagement after immunotherapies using the particular activity of the TCR after binding are currently being explored, including the use of human interleukin-2 (IL-2) cytokine tracers (e.g. ^99mTc-HYNIC-IL-2, ¹⁸F-FB-IL-2)^37-39 and radiolabeled antibodies targeting different TCR domains (e.g. ⁶⁴Cu-cOVA-TCR for quantifying internalization of the TCR-complex⁴⁰ and ⁸⁹Zr-Df-aTCRmu-F(ab’)2 for targeting the beta domain of the TCR⁴¹). In addition to these probes, studies are also exploring the use of PET reporter genes integrated into gene or cell-based immunotherapies^42,43.

Using Downstream Biological Effects to Identify Target Engagement in Molecular Therapies

While directly imaging and quantifying target engagement in human patients remains a significant challenge in brain tumor drug development, biological changes that occur downstream from target engagement can provide significant value in early stages of drug development. For example, studies have suggested target engagement in therapies that inhibit mutant IDH enzymes (e.g. “IDH inhibitors”) in IDH mutant cancers results in reduction in 2-hydroxygluterate (2HG), an oncometabolite detectable using proton MRS that often drives tumorigenesis in these tumors^44,45. Similarly, downstream metabolic and physiologic changes that can be quantified using current imaging technologies may provide value following receptor tyrosine kinase (RTK) inhibitors targeting different oncogenic pathways (Fig. 2). For example, PI3K and mTOR activity is known to play an important role in glucose metabolism^46-49, thus inhibition of mTOR⁵⁰ or RTK-induced activation of PI3K^51,52 can be detected using ¹⁸F-FDG PET. PI3K and mTOR activation also influences angiogenesis⁵³ as well as tumor cell proliferation⁵⁴, suggesting techniques like perfusion and diffusion MRI may provide value in determining adequate target engagement through reduction in vascularity^55-57 and cellularity^58-63, respectively. As an example of this approach, Ellingson et al.⁶⁴ recently used multiparametric MR-PET imaging to explore pharmacokinetics and clinical response to GDC-0084, a brain-penetrant small-molecule inhibitor of PI3K and mTOR. Results from this study showed that composite biomarkers created from ¹⁸F-FDG PET uptake, DCE and DSC perfusion MRI, and diffusion MRI could predict maximum blood concentration, drug exposure, and progression-free survival (PFS) in recurrent GBM treated with GDC-0084, signifying that imaging biomarkers based on downstream effects following target engagement may be useful for early phase drug development studies.

Fig. 2. Example of oncogenic signaling and downstream metabolic and physiologic imaging changes during molecular therapies.

PI3K activation can occur through RAS mutation or increased activation of growth factor receptor tyrosine kinase (RTKs) like EGFR. Activation or mutation of RAS results in stimulation of the RAS/RAF/extracellular signal-regulated kinase (ERK) pathway resulting in increased tumor proliferation. This increase in proliferation can be identified through the use of diffusion MRI, ¹⁸F-fluorothymidine (FLT) PET, or amino acid PET imaging techniques that have been shown to correlate with cellularity and proliferation rates (e.g. ¹⁸F-FET, ¹⁸F-fluorodopa, or ¹¹C-methionine PET). Activation of PI3K/AKT/mTOR pathway after RTK or RAS signaling results in increased angiogenesis through expression of nitric oxide (NO) or through HIF-1α upregulation and secretion of VEGF. To measure changes in vascularity within the tumor, a variety of perfusion-sensitive MRI techniques can be used including dynamic susceptibility contrast (DSC) or dynamic contrast enhanced (DCE) MRI. Changes in oxygenation within the tumor can also be measured using imaging techniques, including oxygen-sensitive R₂’ imaging or radiotracers like ¹⁸F-FMISO that is sensitive to tumor hypoxia. Activation of growth factor RTKs like EGFR also increase glucose metabolism (specific glycolysis) through activation of AKT, MYC, or a variety of other signaling mechanisms. This increase in glucose metabolism can be quantified through the use of ¹⁸F-FDG PET, hyperpolarized ¹³C-glucose or pyruvate, or glucose enhanced CEST imaging (i.e. “glucoCEST”). Lactate and lactic acid production, byproducts of glycolysis, can also be measured using imaging technologies such as pH-sensitive amine CEST MRI, lactate MRS, or hyperpolarized ¹³C-glucose or pyruvate. RTK = receptor tyrosine kinase; MEK = mitogen-activated protein kinase kinase; ERK = extracellular signal-regulated kinase; PI3K = phosphatidylinositol 3-kinase; PIP₃ = Phosphatidylinositol 3,4,5-triphosphate; PIP₂ = Phosphotidylinositol 4,5-Bisphosphate. PTEN = Phosphatase and tensin homolog; PDK1 = Phosphoinositide-dependent kinase 1; eNOS = enzyme nitric oxide synthase; NO = nitric oxide; TSC1/2 = Tuberous sclerosis complex 1/2; RHEB = Ras homolog enriched in brain; mTOR = Mammalian target of rapamycin; mTORC1 = mTOR complex 1; 4E-BP1 = eukaryotic translation initiation factor 4E-binding protein 1; elF4E = eukaryotic translation initiation factor 4E; HIF = hypoxia inducible factor; VEGF = vascular endothelial growth factor; LDH = CEST = chemical exchange saturation transfer; FDG = flurodeoxyglucose; FLT = flurothymidine; FMISO = fluoromisonidazole; MRI = magnetic resonance imaging; MRS = magnetic resonance spectroscopy

Several non-invasive imaging techniques are available for quantifying these downstream effects beyond those previously mentioned. For example, hyperpolarized ¹³C-glucose or ¹³C-pyruvate MRI/MRS can be used to examine metabolic characteristics including glucose uptake, metabolite generation, and lactate production^65-67. Alternatively, investigators recently demonstrated that glucose enhanced MRI can be performed using chemical exchange saturation transfer (CEST) imaging of hydroxyl (-OH) groups on glucose after infusion (glucoCEST)^68-70. Lactate production, a product of glycolysis, can also be quantified using either lactate MRS^71,72 or pH-weighted amine CEST imaging^73-79. For targeted agents that may alter or are influenced by tumor hypoxia⁸⁰, including agents targeting HIF pathways^81,82, carbonic anhydrase IX inhibitors⁸³, agents that directly increase oxygenation⁸⁴, or bioreductive prodrugs that become potent in hypoxic environments⁸⁵, the use of ¹⁸F-fluoromisonidazole (¹⁸F-FMISO) PET^86-89 or oxygen-sensitive MRI^79,90 may be useful. For agents that may alter tumor proliferation rate, diffusion MRI^58-63, ¹⁸F-fluorothymidine (¹⁸F-FLT) PET^91-93, and amino acid PET (e.g. ¹⁸F-fluorodopa⁶⁰) may be advantageous as surrogates of downstream physiologic changes that can be detected using current imaging technologies.

Using Downstream Biological Effects to Identify Target Engagement in Immunotherapies

After initiation of immunotherapies, several molecular, metabolic, and physiologic changes occur downstream from initial target engagement³². Among these downstream changes are alterations in deoxycytidine kinase (dCK) and deoxyguanosine (dGK) within the deoxyribonucleoside salvage pathway. Among the PET tracers being studied as potential biomarkers for dCK activity are 1-(2′-deoxy-2′-[¹⁸F]fluoroarabinofuranosyl) cytosine⁹⁴ (¹⁸F-FAC) and ¹⁸F-clofarabine^95-97 (¹⁸F-CFA), while 2′-deoxy-2′-[¹⁸F]fluoro-9-β-D-arabinofuranosylguanine⁹⁸ (¹⁸F-AraG) is currently being used in a number of clinical trials as a potential biomarker for dGK activity.

In addition to alterations in the deoxyribonucleoside salvage pathway, molecular imaging of amino acid metabolism is another potential useful strategy for using downstream effects to identify target engagement following immunotherapies³². Trans-1-amino-3-[¹⁸F]fluorocyclobutanecarboxylic acid (¹⁸F-FACBC) is currently being studied for use in immunotherapy monitoring and has shown specific uptake in activated immune cells^99,100. Other amino acid PET tracers commonly used for neuro oncology^101,102 also show increased uptake during immunotherapies; however, questions remain as to the specificity of this approach because active tumor also has significant uptake.

Glucose metabolism also appears to be altered in immune cells after activation. Specifically, data suggests elevated glycolytic activity is elevated in GBM exhibiting a favorable immune-stimulatory microenvironment¹⁰³. This suggests that ¹⁸F-FDG PET (and other glycolytic imaging biomarkers including hyperpolarized ¹³C-glucose and glucoCEST) may be useful as a biomarker for immunotherapies¹⁰⁴ as well as molecular or targeted therapies. In solid tumors, data suggests a reduction in ¹⁸F-FDG PET uptake of primary lesions¹⁰⁵ and lack of multiple ¹⁸F-FDG PET avid satellite lesions¹⁰⁶ is suggestive of successful immunotherapy and forms the basis of a new PET Response Evaluation Criteria for Immunotherapy (PERCIMT) criteria¹⁰⁴. However, ¹⁸F-FDG PET uptake can be difficult to interpret in brain tumors due to high activity within active tumor from intrinsic glycolytic signaling as well as uptake in normal brain tissue. Additional studies are needed to better understand the benefits and limitations of using ¹⁸F-FDG PET, hyperpolarized ¹³C-glucose/pyruvate, and glucoCEST to monitor immunotherapy response in brain tumors.

Macrophages, including tumor associated macrophages (TAM) that help create an immunosuppressive tumor microenvironment^107-109, utilize iron as a cofactor for a number of functions including energy production, hypoxic regulation, and inflammation¹⁰⁹. Thus, superparamagnetic iron oxide (SPIO) nanoparticles used as contrast agents in MRI may be useful to get a sense of TAM activity within cancers including brain tumors^110,111. Data suggests accumulation of SPIOs in macrophages occurs 1-10 days after injection¹¹², meaning a patient would need to be dosed at least 24 hours before the imaging exam. Preclinical evidence suggests solid tumors successfully treated with immunotherapies exhibit a reduction in the measured transverse relaxation time constant (T₂ and T₂*)^113,114, implying a reduction in the concentration of TAMs. This reduction in T₂/T₂* also appears to result in a reduction in tumor growth rate¹¹⁵, supporting the use of SPIOs to monitor TAM activity within the tumor. Two commercially available SPIO agents are available for clinical use, ferumoxytol¹¹⁶ (Feraheme, AMAG Pharmaceuticals Inc., Cambridge, MA) and ferucarbotran¹¹⁷ (Resovist, Schering, Berlin, Germany). Ferumoxytol has been used most widely, having the most abundant post-market safety data¹¹² and specifically approved for use in iron replacement therapy, while ferucarbotran was recently approved for use in liver imaging applications but has shown some utility in the CNS¹¹⁸.

Identifying Early Clinical Activity that Increases Likelihood of Meaningful Clinical Outcome

Although overall survival (OS) is the standard for determining treatment efficacy in high grade glioma, early phase clinical trials are typically not powered or designed to identify OS differences. In these early exploratory stages, there could be potential influence of therapies before or after the therapy under investigation. To overcome these limitations and keep study size relatively small, progression-free survival (PFS) and objective response rate (ORR) are considered important end points ¹¹⁹. However, PFS and ORR alone suffer from significant limitations, since the natural history and growth rate for some tumors may make landmark PFS targets (e.g. the rate of PFS beyond 6 months from start of treatment, or PFS6) obtainable by chance or not directly attributed to the treatment under investigation. It is possible additional confidence can be gained by clearly understanding the growth trajectory of a tumor before and after initiating a therapy in addition to landmark PFS targets (Fig. 3). Since previous studies show a clear association between tumor size ^120-129 and growth rates ^120,130-132 with overall survival in GBM, such a framework may be useful for finding clinical activity in early phase trials that increases the likelihood of clinical benefit when advancing to larger and more expensive later stage trials. For example, if a patient has reached the threshold for PFS6 and has a negative or zero (cytostatic) post-treatment growth rate over that period it is more likely that the drug is having meaningful therapeutic activity. If a patient has reached the PFS6 threshold and the tumor growth rate has slowed some, it is possible the drug may be working but there is lower confidence in whether this will translate into meaningful clinical benefit in later stage trials. If a patient reaches the PFS6 threshold but the growth rate after treatment is the same or faster than the pretreatment growth rate, then it is likely the drug is not working. If, on the other hand, a patient does not reach the PFS6 threshold but there is evidence that the drug slowed the tumor growth rate, it is possible there is some therapeutic activity, but the study design or dosing may need to be adjusted in order to translate into clinical benefit. If there is no evidence of extended PFS and tumor growth rate is the same or faster after the treatment, then one can be fairly confident the drug is not working.

Fig. 3. Framework for determining therapeutic activity in early phase trials.

In order to gain confidence that a drug has therapeutic activity that is likely to lead to clinical benefit using a limited dataset in early phase trials, one might consider using landmark progression-free survival (PFS) benchmarks (e.g. PFS6) combined with evidence of altered growth rate trajectory. In patients who reach the PFS benchmark evidence of tumor shrinkage or stabilization would provide confidence the drug is working, whereas if the growth rate has slowed there may be some evidence of activity. Similarly, if patients do not reach landmark PFS but there is evidence the tumor has slowed its growth rate trajectory this may be evidence of therapeutic activity. In these circumstances, adjustments to dose and timing may be appropriate to see clinical benefit.

Use of the modified response assessment in neuro oncology (mRANO) criteria¹³³ may also be useful to gain additional confidence in PFS measurements (Fig. 4). The mRANO criteria was developed as a treatment agnostic criterion for platform and/or adaptive trials where many types of treatments may be compared. The primary difference between the mRANO criteria¹³³ and the standard RANO criteria¹³⁴ is the requirement for confirmation of progression, essentially merging immunotherapy-specific criteria¹³⁵ with prior guidelines. The mRANO criteria has been successfully used in more than a dozen trials to date as an operational tool for managing patients on trial. Recent studies¹³⁶ have also shown that the mRANO criteria is feasible in immunotherapy trials and PFS defined using mRANO appears correlated with OS, providing confidence for use of mRANO-defined PFS as a potential surrogate for clinical benefit in high grade gliomas.

Fig. 4. Modified Response Assessment in Neuro Oncology (mRANO) criteria.

The mRANO criteria¹³³ was developed as a treatment agnostic criterion for use in patient management and to gain confidence in estimates of PFS and pseudoprogression (PsP) incidence. The primary difference between the mRANO and standard RANO criteria is the requirement to confirm progression on a subsequent examination. ORR = objective response rate; PD = progressive disease; SD = stable disease; PsP = pseudoprogression; PR = partial response; CR = complete response.

According to the mRANO criteria, if a patient has no measurable enhancing disease at baseline, then the best response they can obtain is stable disease (SD). However, if enhancing disease becomes measurable in these patients after starting treatment, meaning larger than 1cm x 1cm perpendicular dimensions on at least two slices, then the patient is said to have “preliminary progressive disease (PD)”, and the patient should continue on treatment (if possible) until progression can be confirmed at the next follow up time point. If a patient has measurable disease at baseline, which is mostly the case in studies involving recurrent disease, and after the first post-treatment scan there is no measurable enhancing disease, the patient is said to have “preliminary complete response (CR)”, which then needs to be confirmed at the next time point. If at this next time point (n+1) there is no measurable enhancing or non-enhancing disease, CR is confirmed and the patient stays on treatment until another event occurs. If the patient has non-measurable disease but no enhancing disease, then the patient has a “confirmed partial response (PR)”, and the patient continues on treatment. If the patient has emergence of measurable enhancing disease, then this constitutes preliminary PD and the patient continues on treatment (if possible) until progression can be confirmed at the next follow up time point.

If a patient has measurable disease at baseline and a time point after treatment (n) the lesion shrinks more than 50% from baseline, then the patient has “preliminary partial response (PR)” and the patient continues therapy until the next time point (n+1). If at this time point there is no measurable enhancing disease, this patient has a confirmed PR, but preliminary CR and should continue treatment to see if the preliminary CR can be confirmed at the next follow up time point. Alternatively, if at the confirmation time point (n+1) there remains measurable disease and it is still less than 50% from baseline, this is determined to be a sustained, or confirmed PR and the patient should continue treatment until disease progression is identified. If the lesion is not confirmed to be more than 50% smaller than the baseline and is instead more than 25% larger than the nadir (n) time point, then this constitutes preliminary PD and the patient should continue on therapy until PD can be confirmed at the next time point. If instead the lesion has not grown more than 25% from the nadir but is not 50% smaller than baseline, this constitutes stable disease (and a non-sustained, non-confirmed PR) and the patient should continue treatment until disease progression has been identified and confirmed if possible.

If the patient at any time has a growing lesion that is more than 25% with respect to the nadir, this will constitute preliminary PD and the patient should continue treatment until this can be confirmed at the subsequent time point (n+1). If at this confirmation time point (n+1) the lesion has continued to grow more than 25% from the previous preliminary PD time point (n), then this is confirmed PD and the date of progression is at time (n). If instead the lesion has not continued to grow or doesn’t meet the 25% threshold for confirmation, then this constitutes “confirmed pseudoprogression (PsP)” and stable disease (SD), since it is presumed the lesion has either slowed growth, stabilized, or shrunk with respect to the preliminary PD time point. If PsP is confirmed, then the patient should continue therapy until PD can be confirmed at a later time point (m). If at this later time point (m) the patient has an event in which there is 25% growth with respect to either time point (n), where initial progression was noted, or nadir after PsP at time point (n), then this constitutes confirmation of PD and the date of progression is time (m). If the patient continues to not reach this 25% threshold with respect to either time point (n) or nadir after PsP at time point (n), then this constitutes stable disease (SD) and the patient should continue on treatment until a 2^nd PD event is identified.

In summary, the use of advanced imaging technology can provide a wealth of important information about therapeutic response assessment for both molecular and immunotherapies. These approaches can be tailored to the specific agents being tested and the relevant clinical questions depending on the stage of drug development, ranging from verifying that drug is getting into the brain and hitting the desired target through confirming tumor shrinkage that will increase confidence of a clinical benefit in later stages of testing.