Immunotherapy-related adverse effects on ¹⁸F-FDG PET/CT imaging

Abstract

¹⁸F-Fluorodeoxyglucose positron emission tomography/CT imaging plays a key role in oncological imaging including in staging, radiotherapy planning, treatment response and recurrence assessment. Immunotherapies represent a major advance in cancer therapy for a number of tumours with resulting survival benefit. However, a wide range of immune related adverse events (irAEs), some of which can be apparent on imaging, have been reported. These involve many organ systems but particularly endocrine, cutaneous and gastrointestinal systems. Early detection of irAEs is essential to aid diagnosis and management of patients and to reduce associated morbidity. In addition, it is important to not mistake treatment related effects for disease.

This pictorial review aims to identify common irAEs and changes seen on ¹⁸F-fluorodeoxyglucose positron emission tomography/CT.

Introduction

Immune-modulators are one of the most important classes of new anticancer therapies across a range of tumour types resulting in increased overall and progression-free survival.¹ The Keynote-024 randomised control trial in 305 patients with advanced non-small-cell lung cancer demonstrated significantly improved progression-free survival in patients treated with pembrolizumab compared to standard chemotherapy (10.3 vs 6.0 months respectively).² More recently, combination immunotherapies have also been found to be more effective than individual therapies. The CheckMate 067 trial randomised 945 treatment-naive melanoma Stage III and IV patients into three categories of treatment with individual or combination immunotherapy of ipilimumab and nivolumab. Overall survival at 5 years was demonstrated to be 52% with combination therapy, 44% with nivolumab alone and 26% with ipilimumab alone.³ The side-effects of these treatments vary depending on their mechanism of action.

Immune-related adverse events (irAEs) are more extensively documented with the increasing use of these treatments. Early detection and treatment of these effects is essential for reducing patient morbidity and will help guide changes in subsequent management.

¹⁸F-Fluorodeoxyglucose positron emission tomography (FDG PET)/CT is commonly utilised in staging and response assessment and plays a unique role in detection of inflammatory change especially in the setting of unremarkable CT or MRI imaging. Tumours and inflammation can both have increased glycolysis with increased FDG uptake which may result in interpretive errors. It is therefore essential to recognise common immunotherapy-related changes and be aware of national and international guidance on follow-up, re-assessment and management of irAEs.

Monoclonal antibodies—rituximab

Monoclonal antibodies (mAb) are laboratory produced antibodies against specific/targeted antigens that are expressed on cancer cells.

Rituximab is a mAb to the CD20 protein expressed in B cells and causes cell death through complement-mediated cytolysis and antibody-dependent cell cytotoxicity, which can lead to inflammation and necrosis.⁴ A higher rate of false-positive FDG PET/CT due to inflammatory change has been reported in non-Hodgkin's lymphoma patients receiving cyclophosphamide, doxorubicin, vincristine and prednisolone–rituximab (CHOP-R) compared to CHOP alone.⁴ The false-positive uptake occurs particularly in neck nodes and may be explained by lymphocyte regeneration in peripheral nodes which can be further enhanced by minor infections.

Immune checkpoint inhibitors

Immune checkpoint inhibitors (ICPIs) have become the standard of care for an increasing number of indications, particularly metastatic melanoma, lung cancer and renal cell carcinoma due to increased progression-free- and overall survival benefits in multiple studies.⁵ The most effective classes of ICPIs used in regular oncological practice today are cytotoxic T lymphocyte associated protein-4 (CTLA-4) inhibitors and programmed cell death protein-1 (PD1)/ programmed cell death protein ligand-1 (PD-L1) inhibitors.

CTLA-4 and PD1 are cell membrane proteins that are negative regulators of T cell immune function.⁵ CTLA-4 is expressed on the surface of regulatory T cells; interaction with B7 receptors on antigen presenting cells results in reduction of further T cell activation or immune response expansion.⁶ PD1 is a transmembrane glycoprotein which is expressed on a variety of immune cells. The ligands for PD1: PD-L1 and PD-L2, are found to be more avidly expressed on tumour cells.⁷ PD1-PD-L1 interactions result in down regulation of cytotoxic response by T cells.

The presence of natural inhibitory pathways allows for regulation of the immune system to prevent an autoimmune response. Tumour cells effectively hijack this pathway to limit T cell response and allow tumour cell proliferation. CTLA-4 and PD1/PD-L1 blockade by ICPIs allows the activation and proliferation of T cells, thus restoring the activity of antitumour immune function⁷ (Figure 1).

Figure 1.

Tumour cells dampen T cell response by upregulating inhibition signals from CTLA-4 and PD1 on the T-cell surface. This inhibits T-cell production and allows for tumour proliferation. Checkpoint inhibitors stimulate T cell activation by blocking immune inhibitory checkpoints like CTLA-4, PD1 and PD-L1. This promotes T cell production and restores the anti tumour immune response resulting in tumour cell death by the release of cytolytic molecules, e.g. tumour necrosis factor and interferons. CTLA-4, cytotoxic T lymphocyte associated protein-4; PD1, programmed cell death protein-1; PD-L1, programmed cell death ligand-1.

Recent European Association of Nuclear Medicine (EANM) guidance on FDG PET/CT imaging for response assessment to immunotherapy suggests a baseline FDG PET/CT should be performed with a contrast-enhanced CT and repeated after 2–3 cycles of immunotherapy (usually 8–9 weeks).⁸ In clinical practice, however, response assessment scans are commonly performed after 3–4 cycles (at 9–12 weeks) unless immunotherapy is combined with chemotherapy. In combination therapy, scans are usually performed after 2 and 4 cycles of treatment.

IMMUNE-RELATED ADVERSE EVENTS (irAEs)

Reactivation of the immune system through the use of ICPIs has led to the development of a new series of side-effects. These can involve multiple organ systems including endocrine, musculocutaneous, gastrointestinal, pulmonary and rheumatological. Studies have shown that specific classes of ICPIs will have distinct irAE profiles (Table 1).^8–10 A large systematic review of class-specific patterns of irAEs reported Grade 3–4 irAEs are more common with CTLA-4 inhibitors compared to PD1 inhibitors.⁹ All grades of colitis, hypophysitis and rash were more frequent with CTLA-4 inhibitors whereas pneumonitis, hypothyroidism, arthralgia and vitiligo were more common with PD1 inhibitors. Combination immunotherapy in general is reported to have improved response rates but higher irAEs. In the previously mentioned Checkmate-067 trial, the rate of treatment-related Grade 3–4 adverse events was higher with combination therapy (59%) vs single-agent nivolumab (23%) or ipilimumab (28%).³

Table 1.

Class specific patterns of irAEs⁹

Class of immunotherapy	PD1 inhibitors (e.g. pembrolizumab, nivolumab)	CTLA-4 inhibitors (e.g. ipilimumab)
Response rates	Superior benefit to CTLA-4 inhibitors
Frequency of Grade 3–4 irAEs		More immune related toxicity compared to PD1 inhibitors
Types of irAEs more frequently seen	Pneumonitis, hypothyroidism, arthralgia and vitiligo (PD1 inhibitors > CTLA-4 inhibitors)	Colitis, hypophysitis, rash (CTLA-4 inhibitors >PD1 inhibitors)
Tumour types & irAEs	Melanoma pts higher frequency GI and skin irAEs and lower frequency pneumonitis

CTLA-4, cytotoxic T lymphocyte antigen-4; PD1, programmed death receptor-1; irAEs, immunotherapy-related adverse events.

The importance of recognising irAEs is two-fold: to alert the clinicians to commence appropriate treatment (patients may be asymptomatic at the time) and to prevent misinterpretation of findings as disease progression.

Nodal FDG uptake can be difficult to interpret especially in the context of response assessment as new sites of disease may warrant alteration in the treatment regimen. As seen in Figure 2, it is essential to review the CT component of the PET/CT and the patient's clinical and treatment history to assess if the nature of nodal uptake could be reactive. Ideally, FDG PET/CT should be performed at an appropriate time interval from initiation of treatment to allow for resolution of immediate post-treatment changes.

Figure 2.

63-year-old male with pancreatic ductal adenocarcinoma and bone metastases treated with combination immunotherapy: IMM 101—given as an intradermal injection, gemcitabine and pembrolizumab. (a) Routine follow-up FDG PET/CT to assess for treatment response. MIP image demonstrates intense tracer uptake at the right shoulder tip and within the right axilla. (b) Sagittal fused image of the thoracic spine demonstrating a known bone metastasis (c) Focal intense dermal uptake in the right shoulder corresponds to the site of IMM 101 injection (d) Coronal and axial fused images demonstrating intense FDG uptake within small volume right axillary nodes. The CT component demonstrates normal configuration nodes which are likely to be reactive from the intradermal injection of IMM 101. FDG, ¹⁸F-fluorodeoxyglucose; MIP, maximum intensity projection; PET, positron emission tomography.

Sarcoid-like reaction to treatment has been described following both chemotherapy and immunotherapy.¹⁰ The patient may be asymptomatic with new symmetric mediastinal and hilar nodal enlargement on response assessment imaging and is important to differentiate from metastatic disease (Figure 3). Sarcoid-like lymphadenopathy tends to resolve in most patients.¹¹

Figure 3.

59-year-old female with melanoma of the left forearm. Treated with combination immunotherapy. (a) Initial pre-treatment staging FDG PET/CT MIP image demonstrates avid left axillary nodes (arrow) and low level linear post-surgical uptake in the left upper back. Follow-up FDG PET/CT MIP image following 4 months of treatment demonstrates a partial response to treatment in the left axilla with new moderately avid bilateral mediastinal, hilar and right supraclavicular lymph nodes (circle). (b) Further follow-up FDG PET/CT MIP 4 months following commencement of treatment demonstrates progression of disease at the left axilla with a new left femoral metastasis and left cervical spine metastasis (arrows). The previously demonstrated mediastinal and hilar lymphadenopathy has resolved and was therefore likely to represent benign nodes such as a sarcoid like reaction to treatment. Right iliac fossa activity was felt to be prominent physiological activity.

Reactive splenomegaly (Figure 4) can be seen following treatment with immunotherapy as well as chemotherapy. This can be seen on FDG PET/CT as inversion of the usual liver-to-spleen uptake ratio along with splenomegaly.⁴ This should be monitored to ensure resolution after discontinuation of treatment.

Figure 4.

60-year-old female with urethral melanoma initially treated with radiotherapy. (a) Restaging FDG PET/CT MIP image demonstrates new pulmonary, liver and bone metastases. (b) The patient was commenced on combination ipilimumab and nivolumab therapy. Response assessment FDG PET/CT following 4 cycles of treatment demonstrates a complete metabolic response of the metastases. Also, note a reactive right axillary node and incidental gastric uptake, confirmed to be immunotherapy-induced gastritis on endoscopy. (c) Initial fused coronal PET/CT with avid liver metastases and normal spleen (d) Post-immunotherapy fused coronal image demonstrating complete metabolic response of liver metastases but new increased splenic uptake to liver (inversion of normal liver-splenic ratio) with associated splenomegaly on CT.

Immune-related pancreatitis (Figure 5) is quite rare with an incidence of <1%.¹¹ It is generally associated with a rise in serum amylase but may be clinically asymptomatic. CT and MRI may demonstrate classic features of pancreatitis and PET/CT may show intense FDG uptake.¹⁰ Discontinuation can be considered depending on the severity of presentation.

Figure 5.

50-year-old female, metastatic melanoma from umbilicus with T11 bone metastasis. Treated with combination immunotherapy initially with partial response. Subsequently changed to single agent nivolumab. (a) Pre-nivolumab FDG PET/CT MIP image demonstrates focal uptake in the umbilicus. Incidental avid thyroid nodule was previously assessed with ultrasound + FNA and found to be benign. (b) 3 months post commencement of nivolumab. FDG PET/CT MIP image demonstrates diffuse increased uptake within the pancreas. (c) Fused axial PET/CT image demonstrates intense uptake diffusely through the pancreas. No peripancreatic fat stranding. No history of gallstones or alcohol consumption. This was not associated with inflammatory changes on the CT component, however, serum amylase was identified to be mildly elevated—the appearances are in keeping with a treatment-related pancreatitis.

Thyroid dysfunction (both hypothyroidism and hyperthyroidism) is more common with anti-PD1/PD-L1 treatments.¹² On FDG PET/CT, thyroiditis will commonly result in diffuse homogeneous increased tracer uptake throughout the gland (Figure 6). Small focal areas of uptake require further assessment with ultrasound and possible fine needle aspiration. In the context of thyroiditis, correlation with thyroid function tests (TFTs) is advised. Pituitary hypophysitis (Figure 7) was very rare prior to the introduction of anti-CTLA-4 therapy and was noted in up to 10–13% of patients with melanoma treated with the CTLA-4 inhibitor ipilimumab.¹³ It is a very rare complication in patients treated with PD1/PD-L1 inhibitors.¹⁴ The median onset was 9 weeks after therapy initiation in an ipilimumab cohort.¹⁵ The imaging appearances can be of pituitary enlargement on MRI and/or subtle increased FDG uptake on PET and if the patient is asymptomatic may be found incidentally. Treatment includes systemic high-dose corticosteroids or hormone replacement for secondary hormone deficiencies. Imaging can be used to monitor response to treatment, most commonly contrast MRI.

Figure 6.

81-year-old female with vaginal melanoma treated with radiotherapy and pembrolizumab. (a) Pre-treatment FDG PET/CT MIP image, the primary lesion is not demonstrated on this study. (b) FDG PET/CT MIP Figure 7 weeks post first dose of pembrolizumab. Intense uptake in the thyroid gland and moderate gastric uptake. (c) Axial PET and fused images through the neck demonstrate new diffuse uptake within the thyroid gland. (d) Coronal PET fused images through the abdomen demonstrate uptake within the stomach which was not present on prior imaging. These appearances are in keeping with thyroiditis—thyroid function tests demonstrated a mild degree of hypothyroidism. Gastric uptake is non-specific and could be physiological. In this case, the patient was asymptomatic so gastroscopy was not performed for the low level gastric activity.

Figure 7.

46-year-old male with foot acral melanoma treated with ipilimumab. (a–c) Pre-treatment sagittal FDG PET/CT and T₁W post-contrast MRI imaging appear unremarkable with no evidence of cranial metastastic disease. (d–f) Post 4 cycles of ipilumumab. Sagittal FDG PET/CT and T₁W post-contrast MRI imaging demonstrate focal intense activity arising from an enlarged and avidly enhancing pituitary gland. The appearances are in keeping with pituitary hypophysitis. The patient remained asymptomatic despite imaging appearances and changes resolved following completion of treatment.

The gastrointestinal side-effects of immunotherapy tend to result in enterocolitis and diarrhoea, e.g. rituximab therapy can result in systemic depletion of B cells with severe bowel related adverse events including ileitis and colitis.¹ Diarrhoea and colitis can occur 5 weeks after onset of therapy. Three different patterns of presentation have been reported, diffuse, segmental or isolated rectocolitis.¹⁶ As seen in Figure 8, bowel inflammation can be apparent on FDG PET/CT and foci of bowel mucosal uptake should be carefully reviewed on the CT component for features of inflammatory change, e.g. fat stranding, fluid, free gas or focal collections. However, a common pitfall in diabetics is the presence of metformin therapy related bowel mucosal uptake.⁶ This is typically diffuse but can lead to misinterpretation of long segment FDG uptake. Comparison with previous imaging and correlation with medication history is important to aid interpretation of these findings. In cases where metformin-related bowel activity hampers assessment of suspected bowel pathology consider discontinuation of metformin for subsequent studies ideally for 48 h prior to the PET/CT if diabetic control permits.¹⁷

Figure 8.

59-year-old male with stage T4a N2c M0 left base of tongue SCC. Initially treated with chemoradiotherapy. Follow-up imaging demonstrated bilateral cervical level II nodal recurrence. Subsequently treated with pembrolizumab. (a) Post 14 cycles pembrolizumab FDG PET/CT MIP image. There was a complete response to treatment with no evidence of recurrence on fibreoptic nasoendoscopy. Note diffuse ascending colonic uptake. (b–d) Coronal FDG PET/CT CT components and fused images demonstrate a long segment of mural thickening with mild surrounding fat stranding associated with intense mural FDG uptake. Colonoscopy demonstrated inflammatory change in the caecum and proximal ascending colon. Biopsies showed mild active colitis. The patient was not significantly symptomatic and treatment was continued. At cycle 17, the patient was managing a normal diet with preserved weight. Case provided by Dr Dorothy Gurjal, Consultant Oncologist, Imperial College Healthcare NHS Trust. SCC, squamous cell carcinoma.

Gastritis can be non-specific and not necessarily immunotherapy-related (Figures 4 and 6), however, especially if symptomatic, further evaluation of gastric FDG uptake with endoscopy should be considered.

Myalgia and arthralgia occur more commonly with anti-PD1 therapies with an incidence of 9–12% for patients on pembrolizumab and 6–8% of patients on nivolumab.¹⁸ Development of symptoms should prompt review of recent imaging to assess for signs of inflammation (Figure 9). Treatment can vary depending on severity of symptoms and correlation with signs or symptoms of vasculitis and myositis is required.

Figure 9.

59-year-old male with metastatic melanoma including tiny lung and brain metastases. (a) FDG PET/CT MIP: pre treatment imaging; lung and brain metastases are not easily visualised. (b) 6 months post commencement of treatment with nivolumab: FDG PET/CT MIP demonstrates low level uptake within the left shoulder and both knees. (c) 12 months post-commencement of treatment with nivolumab: FDG PET/CT MIP demonstrates marked FDG uptake within multiple joints including the shoulders, hips, knees and wrists. The appearances are in keeping with an inflammatory bilateral symmetric polyarthropathy. The patient has no past medical history of an arthropathy.

Pneumonitis related to immune-checkpoint inhibitor therapy is relatively rare but can be potentially life threatening (Figure 10). A meta-analysis of 20 clinical trials identified an overall incidence of 2.7% for all-grade pneumonitis.¹⁹ A study demonstrated the median time to development of pneumonitis from treatment initiation to be 2.6 months.²⁰ The spectrum of CT patterns was wide with cryptogenic organising pneumonia as the most common pattern with patchy consolidation and perilobular pattern of ill-defined solid opacities with central ground glass change.²⁰ Development of acute respiratory distress syndrome, non-specific interstitial pneumonia, hypersensitivity pneumonitis patterns have also been described.²⁰ Treatment includes corticosteroid therapy or withholding of immunotherapy depending on the pattern and severity of pneumonitis.

Figure 10.

69-year-old male with metastatic non-small-cell lung cancer treated with nivolumab. (a) Unenhanced CT chest prior to commencement of nivolumab. The primary lesion (not shown) is in the right middle lobe associated with distal atelectasis. A single left hepatic metastasis was also demonstrated. (b, c, e) 6 months post-commencement of treatment with nivolumab: FDG PET/CT axial CT component + fused and MIP images demonstrate new florid subpleural ground glass changes associated with moderate to intense FDG uptake in a non-specific interstitial pneumonia type pattern. The patient was complaining of shortness of breath on minimal exertion and dry cough at the time. The primary mass (not shown) demonstrates reduction in size and low to moderate uptake with stable atelectasis. The patient’s inflammatory markers were not significantly elevated. The appearances were suspected to be immunotherapy-related pneumonitis. Note the single focus of moderate uptake within the left lobe of the liver on the MIP corresponding to the known metastasis. (d) 6 weeks post-corticosteroid therapy, CTPA study performed for sudden onset pleuritic chest pain. This shows significant improvement in the patchy ground glass change. No pulmonary embolus identified. The primary mass (not shown) remains unchanged in size and configuration compared to the PET CT study. CTPA, CT pulmonary angiography.

Interestingly, studies have reported that the development of irAEs caused by immune checkpoint inhibitors is associated with clinical benefit.^21,22 For example, in 148 melanoma patients treated with nivolumab irAEs were observed in 68.2% and a significant overall survival benefit was noted in patients with any grade of irAE vs those without.²¹ Similarly, in patients with advanced non-small-cell lung cancer treated with nivolumab, irAEs were observed in 51%, including 9% Grade 3 or 4 and 18% requiring systemic corticosteroid therapy.²² The median PFS was 9.2 and 4.8 months for patients with or without irAEs, respectively.

Pseudoprogression and immunotherapy response criteria

The novel mechanisms of action of immune modulators may lead to unusual patterns of response that resemble tumour flare but are more pronounced.^23,24 There have been many reports of cases where there has been apparent disease progression but subsequent late deep and durable responses to treatment. This has been termed “pseudoprogression“ and has most commonly been reported in cases of melanoma receiving anti-CTLA-4 treatment, with approximately 15% of patients experiencing pseudoprogression.^25,26 Pseudoprogression is much rarer in other tumour types and with the use of anti-PD1/PD-L1,⁵ therefore, in such cases progression seen on imaging requires careful interpretation to prevent delayed commencement of new lines of treatment.

Pseudoprogression can be seen as an increase in size and avidity of known lesions but also as an increase in number of lesions and development of new sites of disease (Figure 11). Due to the identification of pseudoprogression as a major pitfall in response assessment initially in 2009, modified response criteria based on WHO criteria (including bi-dimensional measurements of up to 10 visceral lesions in total, up to 5 new lesions per organ and 5 new cutaneous lesions) were proposed—the immune-related response criteria (irRC).²⁶ Subsequent studies generally used various RECIST-based immune criteria but mainly as an exploratory end point and it was not standardised. Therefore, a modified RECIST 1.1 for immune-based therapeutics Response Evaluation Criteria in Solid Tumours (termed iRECIST) guidelines have been developed by the RECIST Working Group (Table 2).²⁷ The major change for iRECIST is the concept of resetting the bar if RECIST 1.1 progression is followed at the next assessment by tumour shrinkage. Within this guidance, a new category of unconfirmed progression (iUPD) has been created which requires confirmation of progression (further increase in size or in number of new lesions) by follow up imaging (iCPD). This is generally recommended to be performed between 4 and 8 weeks following the initial study. If progression is not confirmed, but shrinkage occurs (compared with baseline) which meets the criteria of iCR, iPR or iSD, then the bar is reset. If no change in tumour size or extent from iUPD occurs then the timepoint would still be iUPD.

Figure 11.

68-year-old female, metastatic melanoma with lung metastases including a left lower lobe deposit. (a, b) Staging FDG PET/CT prior to commencement of ipilumumab (axial CT and PET images) demonstrates a left lower lobe nodule measuring 10 mm, SUVmax 2.0 (c, d) Follow-up FDG PET/CT 10 weeks later shows mild enlargement of lung nodule with increased FDG uptake, 11 mm SUVmax 3.8 (e, f) Further follow up FDG PET/CT 17 weeks from treatment commencement. This shows interval decrease in size of the lung nodule and reduction in avidity, 4 mm, SUVmax 1.3. These appearances are suggestive of pseudoprogression.

Table 2.

Comparison of RECIST 1.1 and iRECIST²⁷

	RECIST 1.1	iRECIST
Definitions of measurable and non-measurable disease; numbers and site of target disease	Measurable lesions are ≥10 mm in diameter (≥15 mm for nodal lesions); maximum of five lesions (two per organ); all other disease is considered non-target (must be ≥10 mm in short axis for nodal disease)	As per RECIST 1.1, however new lesions are recorded separately on the case report form
Complete response, partial response, or stable disease	Cannot have met criteria for progression before complete response, partial response or stable disease	Can have had iUPD but no iCPD, before iCR, iPR or iSD
New lesions	Result in progression; recorded but not measured	Results in iUPD but iCPD is only assigned if at next assessment additional new lesions appear or an increase in size of new lesions is seen; the appearance of new lesions when none have previously been recorded, can also confirm iCPD
Confirmation of progression	Not required (unless equivocal)	Required
Consideration of clinical status	Not included in assessment	Clinical stability is considered when deciding whether treatment is continued after iUPD

RECIST, Response Evaluation Criteria in Solid Tumours; i, indicates immune responses assigned using iRECIST; iCPD, confirmed progression; iCR, complete response; iPR, partial response.; iSD, stable disease; iUPD, unconfirmed progression.

An adaptation of the Lugano classification for lymphoma response to treatment (Lymphoma Response to Immunomodulatory Therapy Criteria- LYRIC) was created with an “indeterminate response” category to address lesions that may represent pseudoprogression or flare until true disease progression has been confirmed.²⁸

PET based response criteria adapted from PERCIST and EORTC criteria are being explored but have not yet been formally standardised.⁸ For example, one group have suggested PECRIT (PET/CT Criteria for Early Prediction of Response to Immune Checkpoint Inhibitor Therapy) which combines morphological and metabolic criteria in advanced melanoma treated with ipilimumab.²⁹ Another group proposed the PERCIMT criteria (PET response criteria for immune therapy) in a small group of 41 patients.³⁰

Summary

The upregulation of immune pathways through the use of mAbs or ICPIs can result in a wide variety of adverse effects. Some of these will be apparent on FDG PET/CT imaging and it is important to be aware of these to avoid misinterpretation of disease progression from therapy related complications. The utilisation of all components of the PET/CT, prior imaging and the patient's clinical history is essential in the interpretation of response assessment and end of treatment imaging. The development of new treatment protocols and guidance based on multiple clinical trials can also be used to aid interpretation of imaging.

Teaching points

• Reactive adenopathy with increased FDG uptake secondary to vaccinations (immunotherapy or micro-organism/viral) can be misinterpreted as disease progression. Correlation with clinical history is vital.

• Inversion of the normal liver-to-spleen uptake ratio and/or splenic enlargement can be the first sign of immune activity.

• Gastrointestinal and visceral inflammation can be both focal and diffuse, this can be masked by metformin-related mucosal uptake. Close review of foci of increased uptake on both PET and CT is important. Comparison with previous imaging can help assess for subtle changes in uptake or density.

• Thyroiditis, pancreatitis, hypophysitis on FDG PET/CT should be correlated with serum markers. Patients may be asymptomatic but may still require alterations to treatment/steroid therapy.

• Mid treatment scans may show signs of delayed irAEs developing, e.g. arthritis. It is important to compare to prior imaging in order to identify subtle changes.

• Pseudoprogression has been identified in melanoma patients on ipilimumab and is less common with other immune checkpoint inhibitors. Utilisation of iRECIST guidelines when reporting studies which demonstrate possible progression will help advise on timing of follow-up imaging.