Let us start with patient selection. ICI-based therapy plays a key role in standard management of advanced NSCLC, as evidenced by the number of trials conducted in the advanced or metastatic setting to date. The table on this slide highlights several themes common to patient selection across these clinical trials.

First, all of the trials listed here share similarities regarding who was not included in the patient population. When understanding the applicability of clinical trials to real‑world settings, the exclusion criteria can be equally important as—if not more important than—the inclusion criteria. All of these trials excluded persons with a known diagnosis of a comorbid autoimmune condition, especially if that condition were active, meaning the patient had required systemic intervention or treatment in the 2 years before trial enrollment.^1-11 Furthermore, all of these trials excluded patients who were receiving systemic steroids prior to enrollment or those who required concurrent immunosuppressive medications. For the purpose of defining a systemic dose of steroids, multiple trials considered a dose ≥10 mg/day of prednisone or equivalent to be exclusionary.

Second, patients harboring a known oncogenic driver EGFR mutation or ALK rearrangement were also generally excluded from these trials. The notable exceptions were the IMpower150 trial, which did allow patients with these alterations if the patients had experienced progression or unacceptable toxicity with ≥1 approved tyrosine kinase inhibitor.⁸ KEYNOTE‑407 was conducted in patients with squamous NSCLC and did not explicitly exclude patients with EGFR or ALK alterations.⁶ However, current guidelines do not support routine testing for these mutations in patients with squamous histology.¹² Furthermore, EGFR mutations are less frequent and ALK alterations are quite rare in patients with squamous vs nonsquamous NSCLC; hence, it is likely that few patients with these alterations were included in this trial.^13,14 IMpower130 also did not specifically exclude patients with EGFR or ALK alterations, but it did exclude these patients from the primary efficacy and intent-to-treat analyses.⁷

The themes regarding patient exclusion illustrated above in the setting of advanced or metastatic disease have carried through to the design of ICI-based trials in early‑stage NSCLC. Patients with a history of active autoimmune disease and those receiving chronic steroids were also excluded from the IMpower010 and CheckMate 816 trials evaluating adjuvant atezolizumab and neoadjuvant nivolumab in combination with chemotherapy, respectively.^15-19 However, IMpower010 did allow patients with EGFR or ALK alterations to enroll, whereas CheckMate 816 did not.

Clinical trials carefully select participants to balance safety vs efficacy considerations. However, patients in real-world clinics do not always meet the same strict eligibility criteria for clinical trials. Thus, healthcare professionals are faced with the decision whether to offer ICI-based therapy to patients who would have been excluded from the trials leading to the approval of these therapies by the FDA. Perhaps the most common example of one such clinical scenario is whether to offer ICI-based therapy to patients with a history of an autoimmune condition.

Autoimmune conditions are a common phenomenon, with approximately 13.5% of patients with lung cancer having a past or present comorbid autoimmune condition.²⁰ This percentage increases to nearly 25.0% when an alternative definition of “autoimmune disease” is used.

A systematic review of 123 patients with cancer and a known autoimmune condition who received ICI-based therapy has shed some light on safety and efficacy in this population.²¹ It appears that this patient population carries a high risk of immune checkpoint inhibition both exacerbating their known autoimmune condition (which occurred in 50% of patients) and of developing a de novo irAE (in 34% of patients). However, these toxicities were manageable in the vast majority of cases and 50% experienced a response to therapy.

This systematic review also observed that patients who are receiving concurrent immune-suppressing therapy for their autoimmune condition at the start of ICI-based therapy are less likely to develop irAEs (59% vs 83% who were not receiving concurrent therapy). However, data from a retrospective cohort study of 112 patients with autoimmune conditions treated with ICI-based therapy suggest that the median progression‑free survival for those receiving concurrent therapy may be shorter than for those who are not receiving immunosuppressive therapy (3.8 vs 12 months, respectively; P = .006).²²

People with HIV experience an increased risk of cancer compared with the general population, with NSCLC representing the most common non–AIDS-defining malignancy.²³ This increased risk of cancer is likely due to a number of factors, including comorbidities, social factors (eg, smoking, poorer access to preventative services), decreased immunologic surveillance, and increased susceptibility to oncogenic viral infection. Despite this increased risk, the current body of literature is limited on the use of ICI-based therapy in people with HIV. These limited data do suggest that ICIs are both safe and effective in this population, especially when antiretroviral therapy is continued throughout cancer treatment.²⁴ More definitive data will hopefully become available with completion of the ongoing phase II IFCT‑CHIVA2 trial, which is evaluating the safety and efficacy of nivolumab in people with HIV and advanced NSCLC after previous chemotherapy.²⁵

Very limited data are available on the safety of ICI-based therapy for the treatment of cancer in patients who are solid organ transplant recipients, as these individuals were largely excluded from landmark trials of immunotherapy in lung cancer.

A systematic review published in 2020 pooled 57 patients with kidney, liver, or heart transplants who received ICI-based therapy for metastatic malignancies from 2 retrospective studies and 34 case reports or series.²⁶ All of the patients were maintained on ≥1 immunosuppressive agent to prevent graft rejection at the time that ICI-based therapy was initiated. In this review, 37% of the patients experienced graft rejection and 64% of the deaths were due to cancer progression, illustrating the complexities of treating this population while maintaining graft viability.

Another important lesson from a separate review is that graft rejection can occur soon after the start of ICI-based therapy, potentially before any antitumor benefit can be established.²⁷

Now that we have discussed who can be a candidate for ICI-based therapy, we turn to management of irAEs. It is important to understand the unique toxicities that these treatment modalities pose, so that we can quickly identify and manage these events to achieve optimal outcomes for patients.

ICIs modulate the interaction between tumor cells and T-cells by inhibiting tumor cell suppression of T‑cell function—effectively stimulating the immune system.²⁸ This mechanism of action is very different from traditional chemotherapy, which is directly cytotoxic, or targeted therapies such as tyrosine kinase inhibitors, which in general inhibit cancer cell proliferation and/or survival by reducing phosphorylation of receptor tyrosine kinases involved in oncogenesis.^29,30 Furthermore, unlike traditional cancer treatments where cessation of the offending agent often leads to rapid improvement of the toxicity, irAEs can occur up to approximately 2 years after treatment has stopped due to ongoing activation of T-cells.^28,31

To date, there is no biomarker or tool available that can accurately predict who will develop irAEs.³² This means that all patients who will start ICI-based therapy need to be educated on the signs and symptoms of these toxicities. In addition to informing their cancer care team when an irAE arises, patients should be counseled that they must also inform healthcare professionals outside of the oncology team from whom they seek care that they are recipients of ICI-based therapy.³³ This will allow outside healthcare professionals to consider irAEs in their differential when patients seek care for new toxicities or symptoms.

It would be ideal if we could stimulate T-cells in the tumor microenvironment alone. However, ICI-based therapy can stimulate T-cells and cause inflammation in any organ system.^34,35 Uncontrolled inflammation can lead to a host of problems, with the severity depending on the affected organ and the degree of inflammation. The most common irAEs affect the skin, liver, colon, and thyroid, but as shown in the diagram on this slide, many additional organ systems can be affected.

The graph on this slide illustrates the median time to onset of irAEs in patients with NSCLC treated with nivolumab, an anti–PD-L1 ICI.³⁴ As monoclonal antibodies, ICIs are proteins and, therefore, can cause acute infusion reactions either while the patient is undergoing infusion or soon thereafter. These reactions are more likely with the first or second dose and decrease in incidence as treatment continues.

Of the common organ systems affected by irAEs, gastrointestinal toxicities—usually in the form of diarrhea or colitis—are typically the first to occur, whereas pulmonary toxicity or pneumonitis develop later. This should not undermine the need to be vigilant at any point in the treatment course, as toxicities can arise after a single dose or several years of treatment.

As I mentioned earlier, most clinical experience with ICI-based therapy has been in the setting of advanced or metastatic disease. However, patients with early‑stage NSCLC may differ from those with advanced disease in ways that could affect both response to and toxicity from immunotherapy. These factors include lower tumor burden; more robust immune systems because patients are more likely to be treatment naive; potentially higher complication risks from recent surgery; or, in the case of neoadjuvant treatment, irAEs that can complicate the timeline for surgery.³⁶ This raises the question of whether safety outcomes in recent trials of perioperative immunotherapy for early-stage NSCLC were as excepted based on what we know of these agents in the advanced setting.

In the phase III IMpower010 trial comparing adjuvant atezolizumab vs best supportive care after adjuvant chemotherapy in patients with resected stage IB-IIIA NSCLC, safety outcomes were indeed as we might expect.¹⁵ As shown in the table on this slide, approximately 1 in 5 patients (18.2%) discontinued atezolizumab due to AEs. Although 51.7% of patients developed an irAE of any grade, only 7.9% had severe toxicity and 12.1% required the use of systemic corticosteroids. Treatment‑related deaths occurred in only 0.8% of patients in the atezolizumab arm.

Furthermore, the use of adjuvant atezolizumab in the IMpower010 trial led to similar rates, severities, and types of irAEs as those reported in the advanced and metastatic setting.^15,37 Rash, hepatitis, and hypothyroidism were the most common irAEs. Pneumonitis—a particularly important toxicity in the context of NSCLC—occurred at any grade in 3.8% of patients treated with atezolizumab, with grade 3/4 pneumonitis occurring in just 0.8%. That being said, there was 1 death due to pneumonitis in this trial.

Turning now to the neoadjuvant setting, the phase III CheckMate 816 trial is comparing 3 cycles of neoadjuvant nivolumab plus chemotherapy vs 3 cycles of neoadjuvant chemotherapy alone in patients with resectable stage IB-IIIA NSCLC.^18,38 The primary endpoint is pathologic complete response at the time of surgery. As shown in this table summarizing adverse events in CheckMate 816, the addition of nivolumab to 3 cycles of neoadjuvant chemotherapy did not greatly alter the frequencies of treatment‑related adverse events, serious adverse events, surgery‑related adverse events, or treatment‑related deaths.

Among the patients randomized to neoadjuvant nivolumab plus chemotherapy for 3 cycles prior to surgery, there was a low overall incidence of irAEs.¹⁸ The most common irAE was rash, with most of these cases being grade 1/2. Pneumonitis occurred in 1% of patients and all cases were low grade.

Now that we have reviewed the types of irAEs that can occur with ICI-based treatment, we will discuss recommendations for their assessment, monitoring, and evaluation.

Dermatologic assessment requires physical examination as-needed based on patient self‑evaluation.^28,35 It is important to assess the degree of involvement regarding the patient’s body surface area as well as to evaluate for mucosal involvement. Depending on severity, consultation with dermatology may be necessary, especially if a skin biopsy is recommended.

Monitoring for pancreatic irAEs should also follow a symptom‑driven approach. Patients with acute-onset nausea, vomiting, and/or stomach pain should have an assessment of amylase and lipase along with abdominal imaging if pancreatitis is suspected.

Thyroid function testing with thyroid‑stimulating hormone and free thyroxine should be performed every 1‑2 cycles of ICI-based therapy, depending on the interval of dosing. Further laboratory workup may be needed for abnormal results.

Additional monitoring for hypophysitis and other endocrinopathies can occur routinely or with a symptom‑driven approach (eg, if the patient reports excess fatigue). If a serum cortisol test is ordered, it is best drawn in the morning and additional labs may prove helpful based on initial findings; these can include assessing levels of adrenocorticotropic hormone, estradiol, or testosterone.

Regarding pulmonary assessments and monitoring, oxygen saturation levels—especially when treating patients with NSCLC—should be a routine part of vital sign collection when patients check into the infusion center.^28,35 Performing oxygen saturation tests at rest and while walking can help identify patients who may need to undergo an urgent chest CT. Depending on the results of the chest CT, it may be helpful to send the patient for a bronchoscopy with biopsy. Infectious causes can masquerade as suspected pneumonitis on imaging and contribute to shortness of breath and cough.

Cardiovascular complications are among the least common of these toxicities but certainly can be serious. Patients with cardiac comorbidities should be considered for baseline ECG and consultation with the patient’s longitudinal cardiologist. Additional monitoring and follow-up can occur based on symptoms or as recommended by cardiology based on the patient’s condition.

Symptoms affecting the musculoskeletal system and joints can be monitored in patients with preexisting disease. Patients experiencing these events might benefit from referral to a rheumatologist. Additional laboratory testing such as C‑reactive protein, erythrocyte sedimentation rate, and creatinine phosphokinase may be necessary if the clinical situation is worsening.

As is evidenced by this discussion, multidisciplinary support is often necessary to promptly identify and manage irAEs.

Once we have identified and determined the severity of an irAE, we can move onto the potential treatment options.

Generally speaking, patients with grade 1 irAEs can be treated symptomatically and do not require systemic steroids, and treatment with the ICI can continue.^28,35,39 However, the patient will require frequent check‑ins to elicit symptom reporting, as low‑grade irAEs can evolve into more severe toxicities.

For patients with grade 2 irAEs, steroids are needed for those that recur despite previous treatment as well as for select toxicities including colitis, pneumonitis, hepatotoxicity, and renal toxicity. In patients experiencing these select irAEs, prednisone at 1-2 mg/kg/day or equivalent is recommended, and the ICI should be held temporarily until resolution of the irAE. Patients with other irAEs at grade 2 may be able to continue the ICI, except in cases of myocarditis and certain neurologic irAEs.

For patients with grade 3 irAEs, systemic steroids are almost universally mandated and prolonged tapers may be required to control symptoms. The ICI should be held or even discontinued in the case of grade 3 renal toxicity, pneumonitis, or infusion reactions.

Patients with grade 4 irAEs generally require high‑grade systemic steroids to bring the toxicity under control. Again, prolonged tapers over the course of several weeks are usually necessary to control and prevent symptoms from recurring. The ICI should almost universally be discontinued unless this is an endocrine toxicity, in which case replacement therapy can be initiated while the patient is still receiving therapy. For cases of grade 4 irAEs not responding to steroids, additional immune suppressants may be required.

As a pharmacist, it is important to understand how steroids are used in the management of irAEs. Long-term therapy lasting 4-6 weeks may be necessary to prevent their recurrence, as are long tapers lasting at least 30 days.^40-45 To avoid flaring symptoms, rapid tapers are not recommended.

Monitoring of blood glucose is necessary for patients who have preexisting diabetes or are at risk for developing elevated blood glucose due to steroid treatment. Patients receiving prolonged steroids can often develop hypothalamic–pituitary–adrenal axis suppression and need to be monitored for potential adrenal insufficiency. In addition, prolonged steroids can lead to muscle weakness, which can put patients at risk for falls. Thus, physical therapy can be beneficial to patients requiring prolonged steroids.

Other considerations with steroid therapy include the risk of developing gastritis. Patients receiving concurrent nonsteroidal anti‑inflammatory drugs or who also require anticoagulation agents may need to have proton pump inhibitors or H₂ receptor antagonists added temporarily to therapy to prevent the development of gastritis.

In addition, any patient who needs to be maintained on ≥20 mg/day of prednisone for longer than 4 weeks should be considered at risk for opportunistic infections. The principal opportunistic infection of concern is Pneumocystis jirovecii, which requires prophylaxis with sulfamethoxazole and trimethoprim. For patients with sulfa allergies, other medications such as dapsone or atovaquone can be considered. Patients receiving prolonged steroids should also be assessed for risk of fungal infection and may need prophylactic fluconazole.

Osteoporosis is another threat to patients receiving an extended course of steroids. These patients should supplement with calcium carbonate up to 1500 mg/day and vitamin D3 400 IU/day for the duration of the steroid therapy.

For certain irAEs that become refractory to steroids, it is important to administer effective alternative immune suppressants at the first signs of steroid refractoriness.

For patients who develop steroid-refractory grade 2/3 bullous dermatitis, both rituximab and intravenous immunoglobulin (IVIg) have been found to be useful.^38,46,47 Those with steroid-refractory Stevens‑Johnson syndrome or toxic epidermal necrolysis can be treated with IVIg at 1 g/kg/day in divided doses for the course of 3‑4 days to bring the toxicity under control.

For patients who develop steroid‑refractory grade 3/4 elevated transaminases, mycophenolate mofetil is recommended at 0.5‑1.0 g twice daily. It is important to know that the tumor necrosis factor-α inhibitor infliximab is not recommended in the setting of hepatotoxicity as there is a contraindication for the use of infliximab in patients with previous hepatitis.

For steroid‑refractory grade 2-4 diarrhea or colitis, the agents of choice are either infliximab or vedolizumab, an integrin inhibitor. Both agents are administered as 3 doses at Week 0, 2, and 6. Subsequent doses are often not necessary as this dosing interval and schedule has been successful in bringing steroid‑refractory colitis under control. Rarely, patients will require ongoing management with one of these therapies. It is important to note that there have been no head‑to‑head comparisons between these agents; thus, preferential use of one agent over the other is best left to consultation with gastroenterology.

For steroid‑refractory grade 3/4 pneumonitis, several different therapies are available although, again, no head‑to‑head trials have yet been performed. The options are infliximab, IVIg, or mycophenolate mofetil, depending on what is available at the treating institution.

As pharmacists, we want to ensure that our patients with lung cancer receiving ICI-based therapy are educated on the various aspects of their care.

First, the patient should be educated about the goals of ICI-based therapy. In the locally advanced or metastatic setting where treatment is not intended to cure, we want to optimize quality of life, which should place an emphasis on patient symptoms and patient‑reported outcomes. For those patients who are receiving neoadjuvant or adjuvant ICI-based therapy where the goal is indeed cure, it is important for patients to understand the role of therapy in their overall goals. This might mean that patients are educated and encouraged to promptly report symptoms, as these can negatively affect their ability complete therapy or, in the case of neoadjuvant treatment, affect the ability to proceed to surgery with curative intent.

Although patients certainly are not expected to understand all of the mechanistic underpinnings of the immune system, it is important that they acquire a very basic understanding of how ICIs work. This will hopefully help them understand and identify symptoms potentially associated with immune system overstimulation and irAEs.

On a related note, it is critical to reassure patients that the need to use steroids to treat irAEs has not been shown to reduce efficacy of immunotherapy. Indeed, systemic steroids are often vital for the management of severe irAEs and can help optimize outcomes because irAEs that are not controlled can threaten both the organ system and the patient’s life.

Patients also should understand that all of their healthcare professionals, including those outside of their oncology team, must be made aware that they are receiving ICI-based therapy. Toward this end, the Oncology Nursing Society provides a free immunotherapy wallet card summarizing which ICI-based therapy patients are receiving that they can carry and present to any healthcare professional they see.

Finally, patients who are undergoing ICI-based therapy should be counseled to avoid live vaccines during treatment. They can consult the pharmacist or oncology care team for recommendations on the best timing of any necessary vaccinations.