If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Treatment of lung cancer varies according to the histology of the cancer and the presence of targetable mutations and includes conventional chemotherapy, molecular targeted therapy, and immunotherapy.
•
Molecular targeted therapeutic agents and immune checkpoint inhibitors have unique mechanisms of action, which can result in sustained or delayed response, atypical response patterns, and unique treatment-related toxicities.
•
The role of thoracic radiologists in systemic therapy for lung cancer includes characterizing tumor response, identifying early signs of resistance, detecting treatment related toxicity, and planning and performing image-guided biopsies.
Introduction
Non–small lung cancer (NSCLC) has the highest mortality of all malignancies globally. Despite progress and innovations in treatment and management of lung cancer, it remains the leading cause of cancer death. Most patients have advanced incurable disease at the time of diagnosis. Approximately 228,150 new cases of lung cancer and 142,670 deaths related to lung cancer are estimated for the year 2019 [
]. For the purpose of this review, the authors use the term “lung cancer” to refer to malignant neoplasms of the airway and lung parenchyma. They have not included mesenchymal tumors, lymphohistiocytic tumors, tumors of ectopic origin, and metastatic tumors, which constitute the other histologic types of lung neoplasms, as per the 2015 World Health Organization (WHO) classification of lung tumors [
The treatment of lung cancer has evolved dramatically from the early days of conventional chemotherapy, which only resulted in modest survival benefits [
], to the past decade of molecular targeted therapy (MTT) and immune checkpoint inhibitor (ICI) therapy, which has witnessed remarkable improvement in survival [
]. Indeed, the recent strides in the treatment of lung cancer over the last decade have been at the forefront of heralding the current era of personalized cancer therapy. This began with the approval of an epidermal growth factor receptor (EGFR) inhibitor gefitinib in 2003, which was one of the earliest approved MTT agents for solid organ malignancies [
]. The success of this agent prompted the discovery of other targeted therapeutic agents for lung cancer such as the EGFR inhibitor erlotinib in 2004, a vascular endothelial growth factor (VEGF) inhibitor bevacizumab in 2004 [
]. The ICIs nivolumab and pembrolizumab were first approved for lung cancer in 2015 as second-line agents with multiple subsequent studies leading to their approval both as first-line agents and as adjuvant therapy [
Advancements in technology and increased availability of imaging modalities have likewise vastly enhanced the scope of imaging in lung cancer therapy. Imaging modalities such as computed tomography (CT), PET, combined PET-CT, and MRI are now part of the standard guidelines for the management of lung cancer [
]. In this era of precision oncology and personalized care, imaging plays a critical role not only in the detection and initial staging of lung cancer but also in monitoring treatment response, detecting recurrent disease and identifying complications related to disease progression or to treatment.
In this article the authors briefly review the systemic treatment strategies for lung cancer, emphasizing on the differences between the more common NSCLC and the relatively less common SCLC. Subsequently, they discuss in detail the role of imaging in monitoring response and identifying treatment-related toxicities in patients with both NSCLC and extensive stage SCLC receiving systemic therapy. Although both surgery and radiation are key for gaining local control of disease, these are outside the scope of this review.
Guidelines for systemic therapy for lung cancer
Systemic therapy of lung cancer involves use of conventional chemotherapy, MTT, and immunotherapy. Over the past 2 decades, a multitude of targeted therapeutic agents have been introduced for cancer therapy—the “-ibs” that are kinase inhibitors and the “-mabs” that are monoclonal antibodies. Broadly, MTT agents (which include both kinase inhibitors and monoclonal antibodies) are used to block specific receptors and thus limit the growth and survival of tumor cells. ICIs are antibodies that act by blocking mechanisms that tumor cells rely on to evade immune system–mediated damage. Antibodies that target cytotoxic T-lymphocyte–associated protein 4 (CTLA-4) on T cells are termed CTLA-4 inhibitors, whereas antibodies that target programmed death 1 (PD-1) and PD ligand 1 (PD-L1) proteins on tissues are termed PD-1 and PD-L1 inhibitors, respectively. Activation of these proteins leads to downregulation of T cells, and therefore by blocking them, ICIs enable the immune system to destroy tumor cells [
]. An overview of the different drug classes of therapeutic agents and their broad mechanisms of action is shown in Table 1.
Table 1Drugs Used in Systemic Therapy of Lung Cancer
Drug Class
Mechanism of Action
Platinum-based agents (carboplatin and cisplatin)
DNA-binding alkylating agents. Cause intrastrand and interstrand cross-links and thereby damage the integrity of DNA.
Pemetrexed
Inhibits enzymes used in purine and pyrimidine synthesis and thereby prevents the formation of DNA and RNA.
Paclitaxel
Prevents microtubule disassembly, thus blocking the progression of mitosis. This triggers apoptosis or reversal to the G0-phase of the cell cycle without cell division.
EGFR inhibitors
Block EGFR, leading to inhibition of tumor cell growth, survival, and metastasis.
ALK inhibitors
Block ALK receptor, leading to inhibition of tumor cell growth, survival, and metastasis.
ROS-1 inhibitors
Block receptor tyrosine kinases coded by ROS1 gene. This causes inhibition of tumor cell proliferation, survival, and migration.
B-RAF inhibitors
Inhibit mutated B-RAF V600E kinase, leading to reduced signaling through the aberrant mitogen-activated protein kinase (MAPK) pathway. This causes reduced tumor cell growth, proliferation, and survival.
Inhibit receptors for tyrosine kinases coded by fusion genes altering the activity of NTRK domains. This results in inhibition of tumor cell growth, proliferation, angiogenesis, and metastasis.
Immune checkpoint inhibitors
Block cytotoxic T-lymphocyte–associated protein 4 (CTLA-4), PD-1 or PDL-1 receptors. This leads to blockade of the inhibitory signals that enable tumor cells to avoid immune mediated damage.
Epithelial malignancies of the lung can be broadly divided into adenocarcinoma (AC), squamous cell carcinoma (SCC), and large cell carcinoma (LCC). LCC is really a descriptive term to indicate lung cancer with no specific features of SCLC, AC, or SCC. This diagnosis is made on surgical specimens only, whereas its counterpart in biopsy or cytology samples is non–small cell carcinoma not otherwise specified (NSCC-NOS) [
]. Although these 2 terms do not have exactly identical implications, they are interchangeable from a clinical perspective and reflect the inability to identify a specific subtype on histopathology. With the inclusion of immunohistochemistry markers in the 2015 WHO classification of lung tumors, the diagnosis of LCC has decreased, as poorly differentiated carcinomas with AC markers are diagnosed as solid AC and those with SCC markers as nonkeratinizing SCC [
]. The prognosis and management of NSCLC depends on the stage and presence of targetable mutations and therefore only a limited immunohistopathologic analysis is recommended for a relatively smaller biopsy or cytology samples. With such small samples, malignancies that cannot be clearly identified as AC or SCC are labeled NSCC-NOS so as to preserve as much tissue as possible for molecular testing [
Complete surgical resection is recommended for stage I or II disease and stereotactic radiation for those who are not surgical candidates. Adjuvant systemic therapy is indicated for pathologically proven stage II disease and for stage IB disease with high-risk features. Systemic therapy is always indicated for treatment of stage III and stage IV diseases [
The National Comprehensive Cancer Network (NCCN) guidelines mandate testing for presence of targetable mutations in all patients with adenocarcinoma. These include EGFR, ALK, ROS proto-oncogene 1 (ROS-1), B-rapidly accelerated fibrosarcoma kinase (B-RAF), and programmed death ligand 1 (PDL-1) [
] and therefore KRAS screening is not considered mandatory. The NCCN also strongly recommends broad molecular profiling so as to identify rare driver mutations for which targeted drugs may already be available or to give patients the option of participating in ongoing clinical trials for targeted therapeutic agents that act at a molecular level [
Pembrolizumab can be used as a single agent for all NSCLC if PD-L1 expression is greater than or equal to 50%. For PD-L1 less than 50% NSCLC, pembrolizumab is used in combination with conventional chemotherapy. Conventional chemotherapy consists of carboplatin or cisplatin combined with pemetrexed for AC and paclitaxel for SCC. Atezolizumab is an alternative to pembrolizumab for the treatment of AC although it has not yet been approved for SCC [
In select group of patients with AC, bevacizumab is used in combination with conventional chemotherapy, either as initial treatment strategy or as subsequent therapy to be used when the disease develops resistance to the initial therapy [
The treatment options for extensive stage SCLC are unfortunately limited. There are no MTT agents approved for use in treatment of SCLC. The standard treatment regimen uses carboplatin with etoposide[
], and this combination has been in use for many years. Cisplatin may be substituted for carboplatin, although the latter is usually preferred due to less toxicity. Recently, atezolizumab has been approved to be used in conjunction with carboplatin and etoposide as first-line chemotherapy for extensive stage SCLC [
]. SCLC is very responsive to initial treatment although most patients with extensive stage disease develop relapse with resistant disease within 5 to 6 months [
Role of imaging in following disease kinetics and monitoring response to therapy
Imaging plays a pivotal role in monitoring changes in tumor burden in patients on therapy. CT and fluorodeoxyglucose PET-CT (FDG PET-CT) play a complementary role in staging patients with advanced lung cancer. Although the former is important for anatomic characterization, including involvement of critical structures such as airways and mediastinal vascular structures, the latter helps in differentiating metabolically active tumor from surrounding tissues such as atelectatic lung, in identification of small tumor deposits that may not be anatomically apparent, and in guiding biopsy and local treatment.
Although genomic analysis of a tissue sample is essential to characterize the molecular subtype of the tumor, ALK- and EGFR-mutated lung cancers show important differences in clinical and imaging features and also demonstrate different patterns of spread on imaging. For instance, ALK-mutated NSCLC is more likely to present at a younger age and in nonsmokers compared with EGFR-mutated lung cancers. ALK-mutated NSLC tumors are more likely to be central in location, are solid with lobulated contours, and are associated with lymphangitic carcinomatosis. EGFR-mutated lung cancers are more likely to be peripheral in location and are either pure ground glass or part solid nodules with spiculated margins [
Advanced adenocarcinoma of the lung: comparison of CT characteristics of patients with anaplastic lymphoma kinase gene rearrangement and those with epidermal growth factor receptor mutation.
]. The pattern of metastatic spread is also different between these 2 types of tumors. Intrathoracic lymph nodal, hepatic, pleural, and pericardial metastases and sclerotic bone metastases are more likely in ALK-mutated NSCLC, whereas lung, brain, adrenal, and lytic bone metastases are more likely in EGFR-mutated lung cancers [
Advanced adenocarcinoma of the lung: comparison of CT characteristics of patients with anaplastic lymphoma kinase gene rearrangement and those with epidermal growth factor receptor mutation.
MRI is used to detect and follow-up neuroparenchymal metastases, assess epidural extension of spinal osseous metastatic disease, and as a problem-solving tool in specific situations [
]. CT is the most commonly used modality to monitor patients while on therapy, whereas FDG PET-CT is used as a problem-solving tool.
In simple terms, imaging must be able to determine if the tumor burden has significantly changed since the initiation of therapy. Although in the routine clinical practice imaging findings at each scan are always compared with the immediate prior study, it is also essential to identify the scan before start of therapy (baseline) as well as the best response scan (nadir) and compare findings across multiple scans so as to not miss a slowly progressing disease.
Tumor Response Criteria
In order to standardize the assessment of tumor response, the WHO published the first international standard criteria for reporting results of cancer treatment in 1979 [
Personalized tumor response assessment in the era of molecular medicine: cancer-specific and therapy-specific response criteria to complement pitfalls of RECIST.
Personalized tumor response assessment in the era of molecular medicine: cancer-specific and therapy-specific response criteria to complement pitfalls of RECIST.
] as internationally accepted methods for monitoring radiological changes in neoplastic disease while on therapy. The goal of these criteria is to define measurable and nonmeasurable disease and establish cutoffs for progressive disease, stable disease, partial response, and complete response and thereby minimize interobserver variability. The important differences between RECIST 1 and RECIST 1.1 are summarized in Table 2. Currently RECIST 1.1 criteria are used most commonly to monitor treatment response, across cancer types and treatment regimens.
Table 2Differences Between RECIST 1 and RECIST 1.1
Feature
RECIST 1.0
RECIST 1.1
Maximum target lesions
Up to 10 target lesions total, up to 5 per organ
Up to 5 target lesions total, pp to 2 per organ
Measurement of lymph nodes
Not specified
Short axis should be used. ≥15 mm, target lesions; ≥10 mm but <15 mm, nontarget lesions; <10 mm, nonpathologic
Progressive disease
20% increase in longest dimension from nadir
20% increase and >5 mm increase in longest dimension from nadir
PET
Not included
Can be used to support CT or MRI, primarily for progressive disease (detection of new lesions) or confirmation of complete response
Cystic tumors and osteolytic metastases
Not defined
Can be used as measurable disease provided soft tissue component is measurable
Fragmentation or coalescence of target lesions
Not defined
Target lesions that fragment or coalesce can still be measured. The longest dimensions of individual entities are measured in case of fragmentation. In case of coalescence, if it is still possible to identify a separating plane, individual dimensions are used, if not the total longest dimension is used.
]. It is important for radiologists to recognize these features as additional indicators of response, particularly in patients on antiangiogenic agents.
Tumors that respond to EGFR inhibitors tend to develop a scar-like appearance, which makes two-dimensional measurement challenging (Fig. 2). In such patients, measurement of tumor volume has been proposed as an alternate method of monitoring tumor burden [
Fig. 2Scarlike appearance of EGFR mutant lung adenocarcinoma following therapy with EGFR inhibitor. A 47-year-old man with metastatic EGFR mutant lung cancer being treated with erlotinib. The baseline axial chest CT shows a left upper lobe mass with spiculated margins, representing the primary lung cancer (arrow in A). Follow-up imaging 6 months after treatment with erlotinib shows significant decrease in size of the mass, which now has a scarlike appearance (arrow in B), making two-dimensional measurement challenging.
Atypical Response, Pseudoprogression, and Hyperprogression
It is important to recognize that unlike conventional chemotherapy, which works by cytotoxic actions, MTT and ICIs often have a predominantly cytostatic effect. Thus, it is common for patients to remain on these agents for prolonged periods of time with relatively stable overall tumor burden. ICIs have a particularly unique mechanism of action, blocking the inhibitory signals that tumor cells use to avoid attack by immune cells [
]. The resultant infiltration of tumor deposits by immune cells can manifest as an apparent increase in size of the tumor, appearance of new lesions, and increased activity on FDG PET imaging. This phenomenon is referred to as “tumor flare” or “pseudoprogression” [
]. It is important for radiologists and oncologists to recognize this phenomenon and distinguish this from true lack of drug efficacy, as such patients with apparent initial progression of disease may eventually experience meaningful regression (Fig. 3). To account for this unusual pattern of response and avoid premature termination of effective therapy, a new system of tumor response criteria was specifically developed for patients on ICI therapy, namely the immune-RECIST (iRECIST). Besides reformatting the evaluation of patients with pseudoprogression, iRECIST accounts for unconventional mixed or delayed responses. The salient differences between RECIST 1.1 and iRECIST are summarized in Table 3. First, any new measurable or unmeasurable lesions are considered to represent tumor progression in RECIST, whereas in immune related response criteria (irRC) only measurable lesions are included into a composite tumor burden used to evaluate for response or progression. The presence of new unmeasurable lesion in a patient is not necessarily considered to be progression and a short-term follow-up scan is required. Second, with irRC, it is no longer required to have decrease in all measured lesions to be classified as a responder, thus allowing for a mixed response as long as the sum of the measurable lesions or tumor burden is decreasing. Third, progression requires a second imaging evaluation in 4 weeks from the first, to exclude the possibility of a delayed response and distinguish pseudoprogression from true progression.
Fig. 3Pseudoprogression. A 52-year-old man with metastatic lung adenocarcinoma being treated on a clinical trial with nivolumab and citarinostat. At 12 weeks postinitiation of therapy there is increase in size of an anterior mediastinal lymph node on chest CT (arrow in B) compared with pretreatment scan (arrow in A). However, as the patient improved symptomatically, the treatment was continued. A short-interval follow-up scan performed 4 weeks after 3B shows decrease in size of the anterior mediastinal lymph node (arrow in C).
Result in unconfirmed progression. Progression is confirmed on next follow-up if new lesions show increase in size or if additional new lesions are seen. New lesions are not included in sum of target lesions identified at baseline
Response after progression
Cannot be termed as CR, PR, or SD after being labeled as progressive disease
Can have CR, PR, or SD after being labeled as unconfirmed progression
Consideration of clinical status
Not included
Clinical status is considered to decide whether treatment should be continued after unconfirmed progressive disease
A more recently described and poorly understood phenomenon seen with ICIs is hyperprogression. Defining and recognizing this entity is challenging, as it needs measurement of tumor growth rate (TGR), which is not a part of standard tumor response criteria. TGR has been defined as percentage change in tumor volume in 1 month [
]. A greater than or equal to 2-fold increase of TGR between the time point before start of ICI therapy and at first follow-up after initiation of ICI therapy has been defined as hyperprogression [
] (Fig. 4). Hyperprogression has been associated with poor prognosis. In a study by Saada-Bouzid and colleagues, hyperprogression was seen more frequently in patients with regional recurrent disease than with metastases [
], which included several types of malignancies, hyperprogression was not associated with a higher tumor burden. However, in patients with NSCLC, a study by Ferrara and colleagues [
] showed hyperprogression to be associated with a higher metastatic disease burden. Thus, it is not well established if hyperprogression is indeed a distinct entity or merely represents severe progressive disease in patients with an aggressive baseline disease. Further studies are needed to define this entity and develop methods to identify it before treating such patients with ICIs.
Fig. 4Hyperprogression. A 63-year-old woman with metastatic lung adenocarcinoma on nivolumab. Baseline CT abdomen before initiation of therapy shows solitary metastatic lesion in the liver (arrow in A). However, CT abdomen acquired 2 weeks after therapy with nivolumab shows marked increase in size of the previously seen mass with multiple additional mass lesions throughout both lobes of the liver (arrows in B), indicating marked progression of metastatic disease.
Use of MTT agents is associated with an initial dramatic decrease in tumor size, and often there is no measurable disease at the nadir. At this stage detecting even slight increase in disease on imaging is of significance and may indicate development of resistance. Radiologists play a critical role in alerting the treating oncologist of the possibility of developing resistance, as early development of recurrent disease may not have clinical manifestations. Oncologists continue the same treatment for some time even with the development of resistance, as stopping the MTT is associated with a marked rebound increase in tumor burden. This has been referred to as the flare phenomenon [
]. Knowledge of the approximate time for development of resistance to MTT is important, as this helps the radiologists to stay particularly alert to features of progression at the time of expected progression. On an average, NSCLC develops resistance to EGFR inhibitors within 9 to 14 months [
]. Subtle signs of developing resistance include a convex margin in a tumor with previously straight or concave margins, indistinctness of a previously well seen margin, new enhancing focus, new nodular component (Fig. 5) increase in attenuation, filling up of previously patent bronchi (Fig. 6), and increase in solid component or wall thickness of a cavitary lesion.
Fig. 5Early imaging evidence of resistance—new nodular component. A 85-year-old woman with metastatic EGFR mutant lung adenocarcinoma being treated with erlotinib. CT chest performed after 9 months of therapy shows a new nodular component in the posterior aspect of the right upper lobe consolidative mass with slight compression of the adjacent bronchus (arrow in B), compared with the scan done after 7 months of therapy (arrow in A). An axial fused FDG PET-CT performed 1 month after 5B shows increased FDG uptake in the nodule (arrow in C). CT-guided biopsy of the nodule showed a K-RAS mutant lung adenocarcinoma with absence of the original EGFR mutation, indicating development of resistance.
Fig. 6Early imaging evidence of resistance—filling of previously patent bronchi. A 53-year-old man with metastatic lung adenocarcinoma on carboplatin, pembrolizumab, and pemetrexed therapy. Note the patent bronchi within the mixed solid and ground glass mass in the right lower lobe (arrows in A). Follow-up scan after an interval of 2 months shows increase in size of the mass. The previously seen air bronchograms are now nearly obliterated (arrows in B). Also, notable are the diffuse ground glass opacities throughout the right lung, which most likely represents drug-related pneumonitis in this patient presenting with worsening dyspnea and hypoxia.
Image-guided percutaneous biopsy is the definite method of obtaining a tissue sample in patients who are not surgical candidates and in those who cannot have a bronchoscopic biopsy. Obtaining a sample is of utmost importance both to identify the histologic type of the malignancy and to establish a baseline genomic profile. It is the responsibility of the radiologist to determine the safest approach that will also obtain the best yield, and in this regard the radiologist must work closely with the oncologist and pathologist. Fine-needle aspiration (FNA) is associated with a lower overall complication rate. Although, traditionally FNA has been considered to have a lesser yield compared with core biopsy samples, present data suggest that FNA may be adequate in the presence of an experienced pathologist for rapid onsite confirmation [
Improving adequacy of small biopsy and fine-needle aspiration specimens for molecular testing by next-generation sequencing in patients with lung cancer: a quality improvement study at Dartmouth-Hitchcock Medical Center.
]. Sabir and colleagues described an adequacy score in which target factors including smaller size (<2 cm), proximity to high-risk structures, motion, highly angled approach, and tissue factors including sclerosis or necrosis were associated with a higher risk of sample inadequacy [
]. In patients with a predicted higher risk of inadequate percutaneous biopsy sample, radiologists must discuss alternative options with the treating oncologist.
Image-guided biopsy plays an ongoing role during therapy. Multiple repeat biopsies are often necessary to understand the mechanism of resistance, to identify any new targetable mutation and to detect histologic transformation of the neoplasm (Fig. 7) that would need a major change in treatment approach [
]. Multiple biopsies are also often necessary in patients being treated as part of a clinical trial, to monitor effectiveness of therapy.
Fig. 7Transformation of tumor histology while on therapy. A 67-year-old women with metastatic EGFR mutant lung adenocarcinoma treated serially with erlotinib, osimertinib, and combination of carboplatin and pembrolizumab. The baseline CT before start of therapy shows a mass in the right upper lobe representing the primary lung cancer (arrow in A). Restaging CT done 8 months after image A shows marked decrease in size of the mass, which now has a scarlike appearance (arrow in B). Restaging CT done 24 months after A, while the patient was being treated with the third line of therapy, shows increase in size of the right upper lobe nodule (arrow in C). An FDG PET-CT done 3 weeks later shows increased FDG uptake in the right upper lobe nodule (arrow in D) with no significant increase in uptake at other previously known sites of lung adenocarcinoma metastases. (E) A CT-guided biopsy was done to determine the nature of the right upper nodule with the pathology revealing small cell transformation.
It is imperative for radiologists to note that the morphology of the tumor can change markedly with treatment and so repeat biopsy must be carefully planned to target viable tumor and avoid regions of necrosis or scarring.
Role of imaging in monitoring treatment-related toxicity
Drug toxicity continues to be one of the most important challenges in cancer therapy. Radiologists play a critical role in promptly identifying toxicities and conveying them to the treating oncologist. Depending on the severity, adverse effects may warrant cessation or modification of therapy, initiation of mitigating measures such as steroids or hospitalization for supportive care or invasive procedures. The Common Terminology Criteria for Adverse Events (CTCAE) were developed by the National Cancer Institute as a clinical research tool. However, it is now widely used as a standardized method of reporting adverse effects across different types of therapies and toxicities [
The unique mechanisms of action of MTT and ICIs result in equally unique adverse effects. Conventional chemotherapy acts by exerting cytotoxic effects on rapidly dividing cells, and therefore several of their adverse effects are similar, although some agents have specific adverse effects [
]. MTT- and ICI-related toxicities, on the other hand, are intimately related to their actions at the molecular level. Therefore, it is important for radiologists to be fully aware of the specific adverse effects associated with these agents. Table 4 summarizes the important adverse effects seen with MTT and ICI agents used in lung cancer therapy.
Table 4Important Adverse Effects of Molecular Targeted Therapy and Immune Checkpoint Inhibitor Used in Systemic Therapy for Lung Carcinoma
Owing to improved survival due to MTT and ICIs, patients may be treated with these agents for prolonged time periods. This means that radiologists may encounter not only acute toxicities but also delayed and chronic toxicities [
Comparative analysis of immune checkpoint inhibitors and chemotherapy in the treatment of advanced non-small cell lung cancer: A meta-analysis of randomized controlled trials.
]. Detecting such delayed and chronic toxicities can be challenging, as their manifestations are often subtle. Radiologists must remain alert to the possibility of a delayed drug-related toxicity even beyond several cycles of therapy. Often, comparison with multiple prior studies is needed to identify chronic toxicities.
Although ICIs work by blocking the signals that tumor cells use to avoid attack by immune cells, they can often also block such signals in normal cells, resulting in immune-mediated adverse events that may involve multiple organ systems leading to variable manifestations such as pneumonitis, thyroiditis, myocarditis, ocular, neurologic, and hematological manifestations. The severity of the adverse events may be graded according to CTCAE, although it is not tailored specifically to ICI, and management should be supplemented by guidelines released by ASCO and ESMO. Owing to the property of immune system memory, such ICI therapy–related adverse effects may be seen even after stopping the drug, on tapering or stopping steroids—a phenomenon referred to as recurrent toxicity. This has been well described for ICI-related pneumonitis [
] (Fig. 8). However, it is quite likely that any type of ICI-related toxicity may be prone to recurrence.
Fig. 8Recurrent pneumonitis. A 63-year-old woman with metastatic NSCLC on pembrolizumab presented with dyspnea and hypoxia with absence of fever and normal leukocyte counts. (A) Chest CT shows confluent consolidative and ground glass opacities in the right lung thought to represent drug-related pneumonitis. The patient experienced resolution of symptoms after being treated with steroids and discontinuation of pembrolizumab. (B) Chest CT done 1 month after image A shows resolution of the previously seen opacities. The patient was reinitiated on pembrolizumab but rapidly developed worsening dyspnea and hypoxia. (C) Repeat CT at the time of the second presentation shows recurrent consolidative and ground glass opacities in the right lung representing recurrent pneumonitis.
CT findings of chemotherapy-induced toxicity: what radiologists need to know about the clinical and radiologic manifestations of chemotherapy toxicity.
]. Some of the conventional chemotherapy agents used in systemic therapy for lung cancer may also cause pulmonary toxicity. These include pemetrexed, docetaxel, irinotecan, gemcitabine, and temozolamide [
]. Radiologists must consider the possibility of pneumonitis, in the event of unexplained pulmonary opacities, and this is often a diagnosis of exclusion requiring correlation with the clinical picture.
Drug-related pneumonitis has been classified into 5 radiographic patterns based on the American Thoracic Society/European Respiratory Society classifications of idiopathic interstitial pneumonias [
Drug-related pneumonitis during mammalian target of rapamycin inhibitor therapy in patients with neuroendocrine tumors: a radiographic pattern-based approach.
]. Of these patterns, the organizing pneumonia (OP) pattern is the most common and is seen as multifocal consolidative and ground glass opacities that are usually peribronchial and subpleural in distribution (Fig. 9). The reverse halo sign, which refers to an opacity with central ground glass and peripheral consolidation, may be seen with the OP pattern pneumonitis, although this is a nonspecific sign and can be seen with a multitude of causes [
]. The acute interstitial pneumonia/acute respiratory distress syndrome pattern has the highest morbidity and mortality and is seen as diffuse ground glass and consolidative opacities that may be associated with bronchial dilatation (Fig. 10). The hypersensitivity pneumonitis pattern and nonspecific interstitial pneumonia pattern are less common and are associated with lesser severity of pneumonitis. A usual interstitial pneumonia pattern has not been previously described, perhaps because this pattern is associated with the development of pulmonary fibrosis.
Fig. 9Organizing pneumonia pattern of pneumonitis. A 50-year-old woman with metastatic EGFR mutant lung adenocarcinoma on erlotinib presenting with shortness of breath. Chest CT shows multifocal consolidative and ground glass opacities with predominantly peribronchial distribution representing OP pattern of drug-related pneumonitis.
Fig. 10Acute interstitial pneumonia pattern of pneumonitis. A 78-year-old woman with extensive stage small cell carcinoma on combination therapy with nivolumab and ipilimumab presented with dyspnea and was noted to have severe hypoxia. Chest CT shows diffuse ground glass and consolidative opacities throughout both lungs with associated mild bronchial dilatation, representing acute interstitial pneumonia or acute respiratory distress syndrome pattern of pneumonitis.
Imaging as a component of multidisciplinary oncology
In the era of precision oncology, treatment is tailored to the specific needs of the patient, and multiple different treatment regimens are used serially to maximize survival and minimize morbidity [
]. The resultant longer and complex treatment regimens warrant frequent discussions between the oncologist and the radiologist. Although accurately identifying changes in tumor dynamics and treatment-related toxicity are of utmost importance, the final decision to continue or modify treatment is based on a combination of clinical and imaging findings.
Therefore, present practice of oncology mandates an effective system of communication between the oncologist, surgeon, pathologist, and radiologist, who along with other staff involved in the comprehensive care of the patient, constitute the multidisciplinary oncology team [
Systemic therapy for lung cancer continues to evolve with discovery of both new classes of drugs and new-generation agents within existing drug classes that have more specific affinity for the intended receptor and less toxicity. Although NCCN guidelines recommend broad genomic profiling they do not specify which mutations are to be tested for. Individual institutes develop their own panel for genomic testing, and in the future these data are likely to be shared to develop a universally accepted panel. An interesting and evolving tool is the liquid biopsy. This is a test that analyzes circulating tumor cells in the patient’s blood and can potentially obviate true biopsies [
]. Future advancements in this technique may make it much easier to perform genomic testing in all patients and thus lead to identification of more targetable mutations. As per current NCCN recommendations, patients with KRAS mutations are to be treated similar to patients without targetable mutations, unless they can be included in a clinical trial. However, newer agents are being explored to potentially target KRAS mutations [
]. Newer immune checkpoint inhibitors are being explored that use different pathways to enhance immune-mediated damage to tumor cells.
MRI has a limited role in imaging of lung cancer owing to its relatively low spatial resolution and inherent low proton density of lung parenchyma. However, owing to the excellent ability of MRI to characterize soft tissues, it is possible that it may be used to predict and assess response in soft tissue metastases. Studies have shown that low pretreatment apparent diffusion coefficient (ADC) is a predictor of better response for some tumors, and decrease in ADC following therapy is an early indicator of response. Likewise, on perfusion-weighted imaging, higher pretreatment transfer rate of contrast from vascular space into the extracellular space (Ktrans) and higher pretreatment transfer rate of contrast from extracellular space into vascular space (Kep) are associated with better response to antiangiogenic agents. Decrease in Ktrans following therapy is an early indicator of response [
In the rapidly advancing field of molecular imaging, multiple new tracers are being explored. Several studies have reported the utility of probes to intraoperatively detect pulmonary adenocarcinoma nodules, for example, the cetuximab-IRD800, which targets EGFR receptors [
]F-labeled fluoromisonidazole (FMISO). FMISO activity may have prognostic value in predicting response to therapy for NSCLC, with a higher tumor to mediastinal ratio of FMISO activity being associated with a higher risk of relapse at 1 year. Tumors have a higher rate of amino acid transport and protein synthesis, and therefore, there is on-going research to evaluate the ability of tracers such as O-(2–[18F]-fluoroethyl)-l-tyrosine to differentiate inflammation from tumor [
]. Future research must also develop ways to predict toxicities using these tools of quantitative imaging and artificial intelligence.
Summary
In the era of personalized oncology, systemic therapy for lung cancer uses a wide range of tools including conventional chemotherapy, MTT, and immunotherapy. With a broad understanding of the principles of treatment and by using ever advancing imaging techniques, radiologists must strive to effectively monitor treatment response and identify and characterize drug-related toxicities.
Advanced adenocarcinoma of the lung: comparison of CT characteristics of patients with anaplastic lymphoma kinase gene rearrangement and those with epidermal growth factor receptor mutation.
Personalized tumor response assessment in the era of molecular medicine: cancer-specific and therapy-specific response criteria to complement pitfalls of RECIST.
Improving adequacy of small biopsy and fine-needle aspiration specimens for molecular testing by next-generation sequencing in patients with lung cancer: a quality improvement study at Dartmouth-Hitchcock Medical Center.
Comparative analysis of immune checkpoint inhibitors and chemotherapy in the treatment of advanced non-small cell lung cancer: A meta-analysis of randomized controlled trials.
CT findings of chemotherapy-induced toxicity: what radiologists need to know about the clinical and radiologic manifestations of chemotherapy toxicity.
Drug-related pneumonitis during mammalian target of rapamycin inhibitor therapy in patients with neuroendocrine tumors: a radiographic pattern-based approach.
30001-8&id=gr1.jpg){kind=link}
30001-8&id=gr2.jpg){kind=link}
30001-8&id=gr3.jpg){kind=link}
30001-8&id=gr4.jpg){kind=link}
30001-8&id=gr5.jpg){kind=link}
30001-8&id=gr6.jpg){kind=link}
30001-8&id=gr7.jpg){kind=link}
30001-8&id=gr8.jpg){kind=link}
30001-8&id=gr9.jpg){kind=link}
30001-8&id=gr10.jpg){kind=link}