Patient outcomes varied, and were associated with the pattern of histopathologic changes that were present. Clinically, all cases were suspected to represent an adverse reaction to ICI therapy and the ICI agent was discontinued in all cases, with 2 patients receiving supportive care only (one of whom was asymptomatic, and the other having a mild fever only). The remaining 7 patients all received corticosteroids and broad-spectrum antibiotics. Follow-up information was available in 8 patients, 6 of whom had organizing pneumonia with or without granulomas. All 6 of the latter patients did well, with complete resolution of symptoms and imaging abnormalities, and remained without clinical or imaging evidence of pneumonitis at the time of last follow-up (mean: 21.0±15.0 mo later, median: 21 mo, range: 2 to 47 mo). Oncologic responses to ICI therapy were available in 5 of these 6 cases, and included 3 patients with a complete response and no clinical or radiologic evidence of residual malignancy, 1 patient with stable metastatic disease, and 1 patient with slowly progressing disease. The patients with acute fibrinous pneumonitis and DAD both died, 9 days and 8 days after developing symptoms, respectively.
Pneumonitis is a feared complication of ICI therapy and its clinical features are well-documented in the literature.8,9,12–14 Clinical manifestations of ICI-related pneumonitis vary widely, ranging from an insidious onset of respiratory symptoms with subacute or even subclinical asymptomatic disease, to severe and rapidly progressive acute respiratory failure that may be fatal.8 The histopathologic features of ICI-related pneumonitis are less well understood, but it is not surprising that the few available pathologic descriptions in the literature tend to mirror the wide spectrum of clinical manifestations that can occur with these agents. To date, several well-documented single cases have been reported that showed organizing pneumonia,15–17 DAD,18,19 granulomas,19 acute fibrinous and organizing pneumonia,20 and chronic interstitial lymphocytic infiltrates21 as histologic manifestations of this phenomenon, but many patients with clinically suspected ICI-related pneumonitis are never biopsied. Indeed, in one of the largest published series of patients with ICI-related pneumonitis to date, less than half of the patients underwent lung biopsy, and the pathologic findings in these biopsied patients were only available in a small subset.8 Our series corroborates the spectrum of pathologic changes that was reported by Naidoo et al,8 and advances our understanding of this phenomenon by incorporating the clinical and radiologic features in each case. To our knowledge, the present study is the first to systematically detail the histopathologic features of ICI-related pneumonitis and to correlate these findings with clinical and imaging findings, in a series of patients with a variety of malignancies and co-morbidities. Our study is also the first to document foamy macrophage accumulation and pneumocyte vacuolization as essentially universal findings in this context.
Adverse drug reactions remain one of the most challenging diagnoses to establish to a reasonable degree of clinical certainty. In his classic monograph on the subject,22 Irey defined requirements that must be met when a drug reaction is suspected. There must be proof of drug administration, temporal eligibility (ie, exposure to the drug before developing symptoms), and an appropriate latency period between drug administration and development of symptoms. In addition, empiric correlation between the drug and suspected reaction is supported when other possibilities have been eliminated by appropriate clinical and laboratory investigations (eg, a negative workup for infections and autoimmune disorders) or temporal ineligibility (eg, exclusion of other drugs). Empiric correlation is also supported when symptoms improve with discontinuation, when only one drug was given, and/or when the pattern of clinical symptoms is consistent with toxicity. Recurrence of symptoms with reexposure to the drug can further support a putative association, but a drug rechallenge may be unethical and is usually avoided. Using these criteria, 5 degrees of certainty were proposed by Irey (“Causative,” “Probable,” “Possible,” “Coincidental,” and “Negative”). In reality, proving a causal link between a drug and a suspected reaction can be very difficult if not impossible, even after all alternative explanations have been excluded, and cases with the strongest evidence generally fall at most into the “Probable” category. Fortunately, this degree of certainty is sufficient for clinical purposes, to prompt discontinuation of the drug and appropriate therapy.
Like irAEs in other organ systems, ICI-related pneumonitis is a highly unusual type of drug reaction that differs mechanistically from conventional types of adverse drug reactions, likely involving overactivation of the immune system with autoreactive T cells and production of interleukin-17 and other proinflammatory cytokines,1 unlike the various types of hypersensitivity reactions or direct nonimmunologic, cytotoxic, or idiosyncratic reactions that occur with other agents.23 Other potential mechanisms may include modulation of humoral immunity with increasing levels of preexisting autoantibodies, enhanced complement-mediated inflammation due to direct binding of the ICI antibody to CTLA-4 expressed on normal cells, or cross-reactivity with related antigens in normal tissue.1 Whether irAEs should be regarded as true drug reactions in the traditional sense of the term or something else entirely is unclear, but the principles underlying their diagnosis do not depend on the underlying mechanism and Irey’s criteria also apply to irAEs, despite the highly unusual nature of these events. In this study, all of the current cases fulfill Irey’s criteria for ICI agents being a “Probable” cause of the suspected adverse pulmonary reaction. Each patient received ICI therapy before developing symptoms, each showed histologic evidence suggesting an adverse drug reaction (prominent foamy macrophages and pneumocyte vacuolization, often with rare eosinophils), and an appropriate latency period (known to be widely variable with ICI-related pneumonitis8). Each case also fulfills at least one criterion for empiric correlation, including resolution of symptoms or imaging abnormalities with discontinuation, exclusion of infection and other potential causes with an appropriate clinical and laboratory workup, elimination of other drug possibilities, and/or clinical and radiographic findings consistent with a drug reaction. In addition, an ICI agent was the only medication class that was common to all cases. Lastly, 1 patient (case 2) experienced an adverse reaction with pembrolizumab that resolved when the drug was discontinued, followed by a very similar reaction with subsequent nivolumab therapy. These are not identical agents, but they are both directed against the same target (PD-1), and it could be argued that this represents a form of drug rechallenge. We believe these cases fulfill sufficient criteria to suggest a causal link between ICI therapy and the clinical and histopathologic manifestations in the lungs, consistent with Irey’s “Probable” category of certainty.22
As with other drug reactions in the lung, the histopathologic manifestations of ICI-related pneumonitis are nonspecific, and the diagnosis is one of exclusion. Whether organizing pneumonia, granulomas, acute fibrinous pneumonitis, DAD, or other patterns of injury are seen, the differential diagnosis is similar and primarily includes infection, an acute exacerbation or flare of underlying systemic connective tissue disease, and a reaction to some other drug or therapeutic modality (eg, radiation). Clinically, the first and most important distinction to be made is between infection and other causes that may respond to corticosteroid therapy, as this binary decision will influence all subsequent clinical management. It should be remembered that infection always leads the differential diagnosis in the immunocompromised host, and this is often relevant in patients with advanced malignancy and prior chemotherapy. Special stains (eg, GMS and AFB stains or their equivalent) should be utilized liberally to rule in or rule out infection, preferably in combination with microbiologic cultures. Despite the presence of vague granulomas in some cases of ICI-related pneumonitis, necrosis was never seen, and it should be remembered that the probability of infection is high when necrotizing granulomas are present in the lung. If necrosis is seen, other ancillary testing for infectious diseases (eg, serologic and molecular testing) should be considered if cultures and special stains are negative.
Distinguishing de novo ICI-related pneumonitis from pneumonitis related to acute exacerbation of preexisting autoimmune disease is a challenge, as both represent autoimmune phenomena and may be inseparable from a practical standpoint. Fortunately, this distinction may not be critical as both are treated in a similar manner with steroids and immunomodulatory agents if necessary. More important is the question of whether ICI agents can be utilized safely in patients with a preexisting autoimmune disorder. The potential risk of potentiating acute flares of autoimmune disease with ICI therapy is well recognized, but data on the safety and efficacy of ICI agents in these patients is limited. Current evidence suggests that acute flares and irAEs are common in patients with preexisting autoimmune disease, and are usually mild and easily managed and do not necessarily require cessation of ICI therapy, but they can be fatal in rare cases.24,25 In our series, 4 patients had underlying autoimmune disease but none had prior pulmonary manifestations, suggesting that their pneumonitis represented a de novo irAE from ICI therapy, rather than an acute flare of autoimmune disease. Nevertheless, it is certainly possible that these reactions were driven by their underlying immunologic disorders and simply potentiated by ICI therapy.
Distinguishing ICI-related pneumonitis from pneumonitis induced by other drugs or therapeutic modalities is particularly challenging, and is not possible on histologic grounds alone. The list of medications known to cause pulmonary toxicity is long, and many agents can cause DAD, organizing pneumonia, granulomatous inflammation, and other patterns of injury in the lungs. Many agents are also associated with foamy macrophage accumulation, pneumocyte vacuolization, and eosinophil accumulation, and to date, no specific histologic features have been noted with any one agent. Furthermore, foamy macrophages, pneumocyte vacuolization, and eosinophil accumulation are not specific to drug reactions, and can be seen with acute lung injury from a variety of other causes.26–28 Foamy macrophages and eosinophils can be histologic clues that suggest an adverse drug reaction in an appropriate clinical context, but our cases demonstrate that the histopathologic features of ICI-related pneumonitis overlap considerably with those reported with many other drugs and other causes of acute lung injury. Furthermore, drug toxicities can be potentiated by the concomitant administration of other agents or radiation, and it may be difficult to separate the relative contribution of each suspected culprit in a patient who has received multiple medications or other therapies. Indeed, since the early days of ICI clinical trials and continuing to this day, there has been widespread concern about the potential for increased pulmonary toxicity when ICI therapy is combined with radiotherapy and this remains an area of intense investigation, but remains poorly understood.29,30 In our series, several patients received concomitant chemotherapy or radiation, and this is an obvious limitation of our study but is also consistent with a growing body of evidence of increased toxicity with combined immunotherapy and radiation,31 and the potential risk for developing toxicity with combined regimens that include an ICI cannot be ignored. Ultimately, distinguishing ICI-related pneumonitis from pneumonitis caused by some other agent requires a complete medication history, and may only be possible by careful clinicopathologic correlation after a trial of drug discontinuation and clinical follow-up.
Pneumonitis is a potentially life-threatening complication of ICI therapy, but a positive outcome is aided by its early recognition and prompt drug discontinuation. ICI-related pneumonitis is a diagnosis of exclusion requiring high clinical suspicion and sometimes requiring a lung biopsy. The histopathologic manifestations of ICI-related pneumonitis are nonspecific and varied. Biopsies from patients with ICI-related pneumonitis often show a pattern of organizing pneumonia that can be admixed with vague non-necrotizing granulomas in airspaces, and patients with these findings tend to do well, but acute fibrinous pneumonitis or DAD can also occur. In our series, the latter 2 patterns were associated with fatal outcomes. Foamy macrophages and pneumocyte vacuolization are universally seen, and rare eosinophils are frequently encountered. These features are also nonspecific, but their presence may be an important histologic clue that provides additional supportive evidence when ICI-related pneumonitis is suspected. Perhaps even more importantly, in cases where a drug reaction is not clinically suspected, histopathologic features may be the first clue that a patient is experiencing an adverse reaction to an ICI. It would behoove surgical pathologists to be aware of these manifestations to prevent unnecessary delays in the diagnosis, particularly given the rapidly expanding use of ICIs for a variety of indications, and the increasing frequency of irAEs with these agents.
1. Postow MA, Sidlow R, Hellmann MD. Immune-related adverse events associated with immune checkpoint blockade. N Engl J Med. 2018;378:158–168.
2. Wang DY, Salem JE, Cohen JV, et al. Fatal toxic effects associated with immune checkpoint inhibitors: a systematic review and meta-analysis. JAMA Oncol. 2018;4:1721–1728.
3. Borghaei H, Paz-Ares L, Horn L, et al. Nivolumab
versus docetaxel in advanced nonsquamous non-small-cell lung cancer. N Engl J Med. 2015;373:1627–1639.
4. Tay RY, Califano R. Checkpoint inhibitor pneumonitis—real-world incidence and risk. J Thorac Oncol. 2018;13:1812–1814.
5. De Velasco G, Je Y, Bosse D, et al. Comprehensive meta-analysis of key immune-related adverse events from CTLA-4 and PD-1/PD-L1 inhibitors in cancer patients. Cancer Immunol Res. 2017;5:312–318.
6. Khunger M, Rakshit S, Pasupuleti V, et al. Incidence of pneumonitis with use of programmed death 1 and programmed death-ligand 1 inhibitors in non-small cell lung cancer: a systematic review and meta-analysis of trials. Chest. 2017;152:271–281.
7. Nishino M, Giobbie-Hurder A, Hatabu H, et al. Incidence of programmed cell death 1 inhibitor-related pneumonitis in patients with advanced cancer: a systematic review and meta-analysis. JAMA Oncol. 2016;2:1607–1616.
8. Naidoo J, Wang X, Woo KM, et al. Pneumonitis in patients treated with anti-programmed death-1/programmed death ligand 1 therapy. J Clin Oncol. 2017;35:709–717.
9. Suresh K, Psoter KJ, Voong KR, et al. Impact of checkpoint inhibitor pneumonitis on survival in NSCLC patients receiving immune checkpoint immunotherapy
. J Thorac Oncol. 2019;14:494–502.
10. Le T, Minna JD, Gerber DE. Checkpoint inhibitor pneumonitis: too clinically serious for benefit. J Thorac Oncol. 2019;14:332–335.
11. Ma K, Lu Y, Jiang S, et al. The relative risk and incidence of immune checkpoint inhibitors related pneumonitis in patients with advanced cancer: a meta-analysis. Front Pharmacol. 2018;9:1430.
12. Suresh K, Voong KR, Shankar B, et al. Pneumonitis in non-small cell lung cancer patients receiving immune checkpoint immunotherapy
: incidence and risk factors. J Thorac Oncol. 2018;13:1930–1939.
13. Abdel-Wahab N, Shah M, Suarez-Almazor ME. Adverse events associated with immune checkpoint blockade in patients with cancer: a systematic review of case reports. PloS One. 2016;11:e0160221.
14. Suresh K, Naidoo J, Lin CT, et al. Immune checkpoint immunotherapy
for non-small cell lung cancer: benefits and pulmonary toxicities. Chest. 2018;154:1416–1423.
15. Barjaktarevic IZ, Qadir N, Suri A, et al. Organizing pneumonia as a side effect of ipilimumab treatment of melanoma. Chest. 2013;143:858–861.
16. Li H, Ma W, Yoneda KY, et al. Severe nivolumab
-induced pneumonitis preceding durable clinical remission in a patient with refractory, metastatic lung squamous cell cancer: a case report. J Hematol Oncol. 2017;10:64.
17. Kuint R, Lotem M, Neuman T, et al. Organizing pneumonia following treatment with pembrolizumab
for metastatic malignant melanoma—a case report. Respir Med Case Rep. 2017;20:95–97.
18. Shea M, Rangachari D, Hallowell RW, et al. Radiologic and autopsy findings in a case of fatal immune checkpoint inhibitor-associated pneumonitis. Cancer Treat Res Commun. 2018;15:17–20.
19. Koelzer VH, Rothschild SI, Zihler D, et al. Systemic inflammation in a melanoma patient treated with immune checkpoint inhibitors—an autopsy study. J Immunother Cancer. 2016;4:13.
20. Ishiwata T, Ebata T, Iwasawa S, et al. Nivolumab
-induced acute fibrinous and organizing pneumonia (AFOP). Intern Med. 2017;56:2311–2315.
21. Ueno R, Nemoto M, Uegami W, et al. Pembrolizumab
-induced pneumonitis with a perilymphatic nodular pattern in a lung cancer patient: a radio-pathologic correlation. Respir Med Case Rep. 2019;26:168–170.
22. Irey NS. Teaching monograph. Tissue reactions to drugs. Am J Pathol. 1976;82:613–647.
23. Riedl MA, Casillas AM. Adverse drug reactions: types and treatment options. Am Fam Physician. 2003;68:1781–1790.
24. Menzies AM, Johnson DB, Ramanujam S, et al. Anti-PD-1 therapy in patients with advanced melanoma and preexisting autoimmune disorders or major toxicity with ipilimumab. Ann Oncol. 2017;28:368–376.
25. Abdel-Wahab N, Shah M, Lopez-Olivo MA, et al. Use of immune checkpoint inhibitors in the treatment of patients with cancer and preexisting autoimmune disease: a systematic review. Ann Intern Med. 2018;168:121–130.
26. Colby TV. Pathologic aspects of bronchiolitis obliterans organizing pneumonia. Chest. 1992;102:38S–43S.
27. Katzenstein AL, Myers JL, Prophet WD, et al. Bronchiolitis obliterans and usual interstitial pneumonia. A comparative clinicopathologic study. Am J Surg Pathol. 1986;10:373–381.
28. Stanley MW, Henry-Stanley MJ, Gajl-Peczalska KJ, et al. Hyperplasia of type II pneumocytes in acute lung injury. Cytologic findings of sequential bronchoalveolar lavage. Am J Clin Pathol. 1992;97:669–677.
29. Evans T, Ciunci C, Hertan L, et al. Special topics in immunotherapy
and radiation therapy: reirradiation and palliation. Transl Lung Cancer Res. 2017;6:119–130.
30. Wang Y, Deng W, Li N, et al. Combining immunotherapy
and radiotherapy for cancer treatment: current challenges and future directions. Front Pharmacol. 2018;9:185.
31. Verma V, Cushman TR, Tang C, et al. Toxicity of radiation and immunotherapy
combinations. Adv Radiat Oncol. 2018;3:506–511.
Keywords:Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.
immunotherapy; pembrolizumab; nivolumab; drug reaction