Immune-related Adverse Effects Associated with Programmed Death-1 Inhibitor Therapy in the Treatment of Non-Small Cell Lung Cancer: Incidence, Management, and Effect on Outcomes



 

Timothy Chiu PharmD, BCPS1; Christopher Yamamoto PharmD, BCPS2; Fang Niu MS3; Helen Moon MD4; Thach-Giao Truong MD5; Robert Cooper MD4; Rita Hui PharmD, MS6

Perm J 2020;25:20.034 [Full Citation]

https://doi.org/10.7812/TPP/20.034
E-pub: 12/02/2020

ABSTRACT

Background: The programmed death 1 (PD-1) inhibitors may improve survival outcomes of non-small cell lung cancer (NSCLC) patients but are associated with immune-related adverse effects (IRAEs). Management of IRAEs may include immunosuppression (ie, corticosteroids), but there is concern that this may affect efficacy. This study evaluated the influence of IRAEs and immunosuppression for IRAEs on survival outcomes of NSCLC patients treated with PD-1 inhibitors (pembrolizumab and nivolumab).

Methods: We retrospectively examined data from Kaiser Permanente Southern and Northern California members diagnosed with NSCLC who received a PD-1 inhibitor from March 1, 2011 to September 30, 2016. Our primary goal was to evaluate the effects and management of IRAEs on survival with PD-1 inhibitors. Electronic database records were used to identify the occurrence of IRAEs, medication utilization, and death. Cox proportional hazard models were used to evaluate variables for association with increased risk of death.

Results: A total of 662 patients were included in the study (median age = 68 years) (interquartile range 61-74). IRAEs were identified in 18% of patients, of which 62% received immunosuppression. Median overall survival was 10 months (interquartile range = 4 months to not reached). Adjusting for covariates, use of immunosuppression during PD-1 inhibitor treatment was not associated with a significantly higher risk of death (hazard ratio = 1.04, 95% confidence interval = 0.84-1.29), whereas corticosteroid use before initiating PD-1 inhibitor therapy was (hazard ratio = 1.48, 95% confidence interval = 1.14-1.91).

Conclusions: In a large, real-world cohort from an integrated healthcare system, use of corticosteroids prior to PD-1 inhibitors was associated with worse survival outcomes, whereas concomitant treatment was not.

INTRODUCTION

The programmed death 1 (PD-1) inhibitors pembrolizumab and nivolumab reduce inhibition of the body’s innate immune response to unrecognized antigens including cancer cells, resulting in an anticancer effect.1 Both have been approved for the treatment of non-small cell lung cancer (NSCLC) due in part to clinical trial data showing significant increases in median overall survival (OS) of 3 to 4 months compared with standard chemotherapy.2-6

Unlike traditional cytotoxic chemotherapy, these medications are associated with unique and potentially fatal immune-related adverse effects (IRAEs) as a result of impaired self-tolerance from loss of T-cell inhibtion.7 Studies of PD-1 inhibitors have shown IRAEs to include colitis, rash, hepatitis, and other immune-mediated manifestations.2-6 To manage these reactions, protocols have been developed using immunosuppressive agents like corticosteroids in addition to holding or discontinuing PD-1 inhibitor treatment, depending on the severity of the reaction.8-10 There is, however, a theoretical concern that the use of immunosuppressive management (IM) for IRAEs may reduce the therapeutic effect of PD-1 inhibitors by reducing the inflammatory response associated with both IRAEs and treatment efficacy.

Retrospective data from melanoma patients treated with ipilimumab have suggested that the use of IM may not affect survival outcomes.11 Conversely, in pooled analyses of melanoma patients treated with ipilimumab or nivolumab, the occurrence of IRAE has had a neutral to positive effect on survival time.12,13 The influence of IRAEs and IM on the survival outcomes of NSCLC patients treated with nivolumab and pembrolizumab has not been studied in a clinical trial or in a real-world clinical setting.

Thus, the goal of this retrospective cohort study was to evaluate the influence of IRAEs and immunosuppressive treatment of IRAEs on the survival outcomes of NSCLC patients treated with PD-1 inhibitors in an integrated healthcare system based on observational data.

MATERIALS AND METHODS

Health Plan and Oncology Setting

Study Sample

This study was conducted within Kaiser Permanente Northern and Southern California, a large, integrated healthcare delivery system with over 8 million members. Health plan pharmacy databases and cancer registry data were used to identify histologically defined NSCLC patients administered either of the PD-1 inhibitors pembrolizumab or nivolumab during March 1, 2011 to September 30, 2016. All patients were presumed to have received cancer treatment within Kaiser Permanente due to the Kaiser Permanente’s structure as a health maintenance organization. Index date was defined as the first day of PD-1 inhibitor administration, and follow-up continued until patient death; end of Kaiser Permanente Membership; or December 31, 2016; whichever came first. This study was approved by the Kaiser Permanente Northern and Southern California Institutional Review Boards, and a waiver of informed consent was obtained due to the nature of the study.

Outcomes and Covariables

Medical, laboratory, and pharmacy dispensing data were collected from the Kaiser Permanente California Region electronic database. Baseline characteristics included patient age, sex, ethnicity, comorbidities, and corticosteroid use recorded up to 6 months before index. Corticosteroid use before index was converted to a cumulative prednisone equivalent dose over the period of 6 months before index date. Comorbidities used to calculate Charlson Comorbidity Index included data up to 1 year before index, with the calculation of the Charlson Comorbidity Index performed as described in the medical literature.14

Cancer type was identified using histological data from the Kaiser Permanente Northern California Cancer Registry (KPNCCR) and the Kaiser Permanente Southern California Cancer Registry (KPSCCR). KPNCCR and KPSCCR were established in 1994 and 1998, respectively. Both registries manage a database of all Kaiser Permanente California members with a cancer diagnosis. The KPNCCR currently includes cases diagnosed from 1947 to present, and the KPSCCR includes cases diagnosed from 1980 to present. Both registries document invasive and in situ cancer, including all solid tumors (eg, breast and prostate) and systemic malignancies (eg, lymphoma, leukemia, and multiple myeloma). These cases are reported, as mandated, to the California Cancer Registry and the national Surveillance, Epidemiology, and End-Results Program, which prepares an annual report.

IRAEs were identified by International Classification of Diseases, 9th or 10th edition, Clinical Modification (ICD-9 or ICD-10) codes for a variety of different conditions deemed by investigators to be related to IRAEs (Table 1). Certain diagnoses for conditions considered to be chronic were excluded if the patient had a previous diagnosis for the same condition up to 1 year prior to index date. Diagnoses were considered to be related to IRAEs if they occurred after index and up to 180 days after the last PD-1 prescription dose.

IM was defined as the use of any systemic corticosteroid or alternate immunosuppressive medication used up to 30 days after an IRAE. Systemic corticosteroids included oral or injected betamethasone, cortisone, dexamethasone, hydrocortisone, methylprednisolone, prednisone, prednisolone, and triamcinolone. Alternate immunosuppressive medications included adalimumab, antithymocyte globulin, cyclophosphamide, cyclosporine, infliximab, mycophenolate, and tacrolimus. IM use in patients without an identified IRAE during PD-1 inhibitor treatment and up to 30 days after the end of PD-1 inhibitor treatment was also recorded and analyzed.

Date of death was determined from Kaiser Permanente Beacon and Kaiser Permanente inpatient records and membership data.

Statistical Analysis

OS was compared between groups defined by IRAE occurrence, IM, and whether subsequent doses of PD-1 inhibitor were administered after the date of IRAE (Figure 1). OS outcomes were compared using Kaplan-Meier estimates, with log rank tests used to compare differences between groups. Cox proportional hazard modeling was used to determine the relative risk of death associated with various baseline and treatment variables. Additional post hoc sensitivity analyses were performed on different criteria for the definition of preindex corticosteroid use by varying the parameters of the variable entered into the model. A p value less than 0.05 was used as the criterion for statistical significance. All data were analyzed using SAS version 9.4 (SAS Institute, Cary, NC).

Figure 1

Figure 1. Immunosuppressive management of immune-related adverse effects in non-small cell lung cancer patients treated with programmed death receptor-1 inhibitors. DC = discontinued; IM = immunosuppressive management; IRAE = immune-related adverse effect.

Figure 2

Figure 2. Unadjusted survival outcomes of non-small cell lung cancer patients treated with programmed death receptor-1 inhibitors stratified by occurrence of immune-related adverse effect, immunosuppressive management, and continuation or discontinuation of therapy after immune-related adverse effect. DC = discontinued; IRAE = immune-related adverse effect; IS = immunosuppression.

RESULTS

Cohort Characteristics

A total of 662 patients were included in the study cohort (Table 2). Median follow-up time was 7 months (interquartile range [IQR] = 4-11 months). During the study period, there were 349 deaths, with a median time from drug exposure to death of 10 months (IQR = 4 months to not reached); 93% of patients were treated with nivolumab versus 7% with pembrolizumab, with a median PD-1 exposure time of 3 months (IQR = 2-6 months).

IRAE Incidence

IRAEs were identified in 114 (17%) of patients at a median time to occurrence of 64 days (IQR = 27-126 days) (Figure 1). Dermatologic, gastrointestinal, and pulmonary IRAEs were the most commonly reported at 8%, 3%, and 3%, respectively. Timing and incidence of IRARs were detailed in Table 3.

IRAE Management

Of the 548 patients not identified as having an IRAE, 259 (47%) received IM between index and 30 days after the end of PD-1 inhibitor treatment. Of the 114 patients with identified IRAEs, 71 (62%) received IM within 30 days of the IRAE, but the remaining 43 (38%) did not. Of the 71 patients treated with IM, 49 (69%) did not receive subsequent PD-1 inhibitor doses after IRAE, whereas 20 of 43 (47%) patients who did not receive IM did not receive subsequent PD-1 inhibitor doses after IRAE.

Overall Survival

The median survival time for the cohort was 10 months (IQR = 9-11 months). Median survival time in the groups that continued therapy after the occurrence of an IRAE was numerically higher than in the groups with no IRAE. Conversely, the median survival time for groups that discontinued therapy after IRAE was lower than groups with no IRAE. Groups treated with IM had similar survival times compared with those without IM (Table 4). Kaplan-Meier curve was shown in Figure 2.

Cox proportional hazard modeling of mortality risk showed that IM during PD-1 inhibitor treatment was not associated with a significant increase in mortality risk (hazard ratio [HR] = 1.04, 95% confidence interval [CI] = 0.84-1.29) (Table 5). In contrast, the use of corticosteroids up to 6 months before initiation of PD-1 inhibitor treatment was associated with a significantly increased mortality risk (HR = 1.48, 95% CI = 1.14-1.91). Compared with patients with no IRAE, the occurrence of IRAE followed by continued therapy was associated with a significantly reduced mortality risk (HR = 0.49, 95% CI = 0.30-0.82), whereas an IRAE followed by treatment discontinuation was associated with a significantly higher mortality risk (HR = 1.37, 95% CI = 1.01-1.86). None of the other variables included as part of the Cox model was a significant predictor of mortality, except for Charlson Comorbidity Index equal to or greater than 5 (HR = 1.36, 95% CI = 1.00-1.85).

Sensitivity analyses were performed to further investigate the influence of corticosteroids used before PD-1 inhibitor initiation, which included the effects of different time and dose criteria (Table 6). All criteria for time of corticosteroid use from 2 weeks to 6 months prior produced significantly increased risk of death with corticosteroid exposure before initiation of a PD-1 inhibitor. Variation in dose criteria showed a significantly increased risk of death when no dose criteria were used and with a cumulative prednisone equivalent dose of 1000 mg or greater. Risk of death was numerically higher but not statistically significant with prednisone equivalent doses of 500 mg or greater.

Table 1. International Classification of Diseases Diagnosis Codes for immune-related adverse effects

Category Excluded if preexisting* ICD-10 ICD-9 Description
Dermatologic No L139 6949 Bullous dermatitis
Dermatologic No L27 - Dermatitis due to substance taken internally
Dermatologic No L270 6930 Dermatitis due to substance taken internally
Dermatologic No L271 6930 Dermatitis due to substance taken internally
Dermatologic No L278 6938 Dermatitis due to substance taken internally
Dermatologic No L279 6939 Dermatitis due to substance taken internally
Dermatologic No L29 - Pruritus
Dermatologic No L298 6988 Pruritis other
Dermatologic No L299 6989 Pruritis NOS
Dermatologic No L309 6929 Dermatitis
Dermatologic No L309 - Drug reaction/eruption
Dermatologic No L51 - Erythema multiforme
Dermatologic No L511 69,513 SJS
Dermatologic No L512 69,515 TEN
Dermatologic No L513 69,514 SJS/TEN overlap
Dermatologic No L518 69,511 Other erythema multiforme
Dermatologic No L518 69,512 Other erythema multiforme
Dermatologic No L518 69,519 Other erythema multiforme
Dermatologic No L519 69,510 Erythema multiforme unspecified
Dermatologic No L80 70,901 Vitiligo
Dermatologic No - 6929 Dermatitis
Dermatologic No - 9952 Drug reaction/eruption
Dermatologic No - 502,164 Dermatitis due to chemotherapy
Dermatologic No R21 7821 Rash NOS
Endocrine No E032 2443 Hypothyroidism due to medicaments and other exogenous substances
Endocrine Yes E038 2448 Other specified hypothyroidism
Endocrine Yes E039 2449 Hypothyroidism unspecified
Endocrine Yes E0580 24,280 Other thyrotoxicosis
Endocrine Yes E0581 24,281 Other thyrotoxicosis
Endocrine Yes E0590 24,290 Thyrotoxicosis, unspecified
Endocrine Yes E230 2532 Hypopituitarism
Endocrine Yes E230 2533 Hypopituitarism
Endocrine Yes E230 6281 Hypopituitarism
Endocrine No E231 2537 Drug-induced hypopituitarism
Endocrine Yes E236 2534 Other disorders of pituitary gland
Endocrine Yes E236 2538 Other disorders of pituitary gland
Endocrine Yes E237 2539 Disorder of pituitary gland, unspecified
Endocrine Yes E271 25,541 Primary adrenocortical insufficiency
Endocrine No E273 - Drug-induced adrenocortical insufficiency
Endocrine Yes E2740 25,541 Unspecified adrenocortical insufficiency
Endocrine Yes E2749 25,542 Other adrenocortical insufficiency
Endocrine Yes E2749 2555 Other adrenocortical insufficiency
Endocrine Yes E278 2558 Other specified disorders of adrenal gland
Endocrine Yes E279 2559 Disorder of adrenal gland, unspecified
Gastrointestinal No K2900 53,500 Acute gastritis
Gastrointestinal No K2901 53,501 Acute gastritis
Gastrointestinal No K2960 53,520 Other gastritis
Gastrointestinal No K2961 53,521 Other gastritis
Gastrointestinal No K2970 53,550 Gastritis unspecified
Gastrointestinal No K2971 53,551 Gastritis unspecified
Gastrointestinal No K2980 53560 Duodenitis
Gastrointestinal No K2981 53,561 Duodenitis
Gastrointestinal No K521 5582 Toxic gastroenteritis and colitis
Gastrointestinal No K5289 5589 Noninfective gastroenteritis and colitis
Gastrointestinal No K529 5589 Enterocolitis
Gastrointestinal No - 78,791 Diarrhea
Gastrointestinal No - 53,540 Other gastritis
Gastrointestinal No - 53,541 Other gastritis
Gastrointestinal No - 5368 Gastritis/dyspepsia
Gastrointestinal No R197 - Diarrhea
Hepatic Yes K71 - Toxic liver disease
Hepatic Yes K716 5733 Toxic liver disease
Hepatic Yes K72 - Hepatic failure NOS
Hepatic Yes K7200 570 Hepatic failure NOS
Hepatic Yes K7210 5728 Hepatic failure NOS
Hepatic Yes K7290 5722 Hepatic failure NOS
Hepatic Yes K754 57,142 Autoimmune hepatitis
Hepatic Yes K759 5733 Inflammatory liver disease NOS
Neurologic Yes G7000 35,800 Myasthenia gravis without (acute) exacerbation
Neurologic Yes G7001 35,801 Myasthenia gravis with (acute) exacerbation
Neurologic No I6783 34,839 Posterior reversible encephalopathy syndrome
Neurologic No G0481 32,381 Other encephalitis and encephalomyelitis
Neurologic No G0489 32,382 Other myelitis
Neurologic No G0489 3239 Encephalitis and encephalomyelitis, unspecified
Neurologic No G0491 3239 Myelitis, unspecified
Pancreatic No K853 - Drug induced pancreatitis
Pancreatic No K858 - Other acute pancreatitis
Pancreatic No K859 5770 Pancreatitis
Pulmonary No J702 - Acute drug-induced interstitial lung disorders
Pulmonary Yes J703 - Chronic drug-induced interstitial lung disorders
Pulmonary No J704 - Drug-induced interstitial lung disorder, unspecified
Pulmonary No J708 5088 Respiratory conditions due to external agents
Pulmonary No - 486 Pneumonitis
Renal Yes N052 5831 Unspecified nephritic syndrome
Renal Yes N055 5832 Unspecified nephritic syndrome
Renal Yes N058 58,389 Unspecified nephritic syndrome
Renal Yes N059 5830 Unspecified nephritic syndrome
Renal Yes N2889 59,389 Unspecified nephritic syndrome
Renal Yes N289 5939 Disorder of kidney and ureter, unspecified
Renal Yes - 5834 Unspecified nephritic syndrome
Renal Yes - 5839 Unspecified nephritic syndrome
Renal Yes - 58,381 Nephritis

*Excluded if same diagnosis code was recorded up to 1 y before index.

NOS = not otherwise specified; SJS = Stevens-Johnson syndrome; TEN = toxic epidermal necrolysis.

Table 2. Baseline characteristics (n = 662)

Number of patients 662
Median age at treatment initiation, IQR 68 (61-74)
Male, % 51
Ethnicity, %
 White 61
 Asian/Pacific Islander 17
 African American 12
 Hispanic 10
 Other 1
Median Charlson comorbidity index14 3
Chronic pulmonary disease, % 46
Use of corticosteroids before starting PD-1 inhibitor, %
 Within 6 mo  
  Any dose 73
  ≥ 500 mg prednisone equivalent 17
  ≥ 1000 mg prednisone equivalent 1
 Within 3 mo 56
 Within 2 mo 47
 Within 1 mo 26
 Within 2 wk 13

PD-1 = programmed death 1. mo = month.

Table 3. Incidence and timing of immune-related adverse effects in non-small cell lung cancer patients treated with programmed death receptor-1 inhibitors (n = 662)

Dermatologic  
 % 8
 Median days to occurrence, n (IQR) 81 (35-132)
Gastrointestinal
 % 4
 Median days to occurrence, n (IQR) 83 (43-154)
Pulmonary
 % 3
 Median days to occurrence, n (IQR) 31 (11-65)
Hepatic
 % 2
 Median days to occurrence, n (IQR) 102 (52-201)
Renal
 % 1
 Median days to occurrence, n (IQR) 55 (28-108)
Pancreatic
 % 0
 Median days to occurrence, n (IQR) 52 (13-91)
Endocrine
 % 1
 Median days to occurrence, n (IQR) 69 (29-90)
Neurologic
 % 0
 Median days to occurrence, n (IQR) 176 (176-176)
Any IRAE
 % 17
 Median days to occurrence, n (IQR) 64 (27-126)

Twelve patients (2%) experienced >1 type of immune-related adverse effect.

IQR = interquartile range.

Table 4. Unadjusted overall survival outcomes of non-small cell lung cancer patients treated with programmed death receptor-1 inhibitors

IRAE + Continue (n = 23)
 Median months survival NR
 95% CI 12-NR
IRAE + IS + Continue (n = 22)
 Median months survival 17
 95% CI 11-NR
IRAE + DC (n = 20)
 Median months survival 6
 95% CI 2-NR
IRAE + IS + DC (n = 49)
 Median months survival 6
 95% CI 4-9
No IRAE (n = 289)
 Median months survival 10
 95% CI 8-13
No IRAE + IS (n = 259)
 Median months survival 10
 95% CI 8-11

DC = discontinued; IM = immunosuppressive management; IRAE = immune-related adverse effect; NR = not reached.

Table 5. Cox-proportional hazard model of risk associated with various baseline and treatment factors in non-small cell lung cancer patients treated with programmed death receptor-1 inhibitors

  Overall survival
Factor HR 95% CI p value
IM during PD-1 inhibitor treatment: yes vs no 1.04 0.84-1.29 0.71
PD-1 after IRAE occurrence: yes vs no IRAE 0.49 0.30-0.82 <0.01
PD-1 after IRAE occurrence: no vs no IRAE 1.37 1.01-1.86 0.05
Corticosteroid use before PD-1 inhibitor treatment: yes vs no 1.48 1.14-1.91 <0.01
Charlson Comorbidity Index: 5 vs 1-2 1.36 1.00-1.85 0.05
Charlson Comorbidity Index: 3-4 vs 1-2 1.29 1.00-1.67 0.05
PD-1 inhibitor: pembrolizumab vs nivolumab 1.25 0.83-1.88 0.29
Ethnicity: white vs non-white 1.09 0.87-1.37 0.44
Sex: male vs female 0.98 0.79-1.21 0.82
Age: ≥65 y vs <65 y 0.84 0.66-1.06 0.13

CI = confidence interval; HR = hazard ratio; IRAE = immune-related adverse effect; PD-1 = programmed death 1.

Table 6. Sensitivity analysis of altering parameters of corticosteroid use before programmed death receptor-1 inhibitor initiation

Time (any dose)
 6 mo  
  HR 1.48
  95% CI 1.14-1.19
 3 mo  
  HR 1.39
  95% CI 1.12-1.73
 2 mo  
  HR 1.43
  95% CI 1.16-1.77
 1 mo  
  HR 1.37
  95% CI 1.09-1.73
 2 wk  
  HR 1.87
  95% CI 1.41-2.49
Dose (over 6 mo)
 Any dose  
  HR 1.48
  95% CI 1.14-1.91
 500 mg cumulative prednisone equivalent  
  HR 1.3
  95% CI 0.99-1.69
 1000 mg cumulative prednisone equivalent  
  HR 2.38
 95% CI 1.08-5.21

CI = confidence interval; HR = hazard ratio. mo = month.

DISCUSSION

To our knowledge, the current study is the largest single retrospective study of the outcomes of NSCLC patients treated with PD-1 inhibitors within the United States population. We found IRAE incidence and management rates that were comparable to those seen in clinical trials. Our data also suggest that, although the use of corticosteroids before PD-1 inhibitor treatment initiation and treatment discontinuation after IRAE are associated with worse OS outcomes, the use of IM during PD-1 inhibitor treatment was not.

The rates of IRAE and IRAE management seen in this study are difficult to compare with those found in existing literature due to differences in study methodology. The vast majority of data available on the adverse effect profile of pembrolizumab and nivolumab comes from the clinical trial setting.2-6 In this highly regulated and structured environment, the level of adverse effect detection and reporting are likely to be different than what we have seen in our real-world retrospective study. Furthermore, our methodology using ICD codes to identify IRAEs lacks severity information, and thus it is difficult to determine how our IRAE detection compared with clinical trials. Treatment-related adverse effects were reported in clinical trials at rates of 58% to 73%, with rates of grade 3 or higher adverse effects at 7% to 27%.2-6 Specifically, KEYNOTE-010 and KEYNOTE-024 reported rates of 19% to 29% for investigator-designated immune-mediated adverse effects.5,6 These numbers are similar to the 17% identified in our current study, although differences in detection undoubtedly exist between our retrospective methodology and the individually assessed adverse effects from clinical trials. Overall, the trends in IRAE incidence seem comparable to our study. Gastrointestinal and dermatologic effects appear to be relatively more common, and the median time to occurrence of IRAE appears to be roughly 1 month to 2 months.

Regarding the use of IM to manage IRAEs, the authors of the CheckMate-017 and CheckMate-057 studies reported steroid use rates of 11% to 83% in NSCLC patients treated with nivolumab depending on the type of IRAE.2,3 These numbers are in alignment with the overall 62% seen in the current study. It is also interesting to note that almost half of the patients in our study with no identified IRAE required IM for another indication. Preliminary chart review has shown that these indications included chronic obstructive pulmonary disease exacerbations, brain metastases, and procedural preparation; however, these patients warrant further formal investigation.

Treatment discontinuation rates found in previous literature ranged from 3% to 7.1%, whereas the proportion of patients in our study who did not receive subsequent doses of PD-1 inhibitors after an identified IRAE was 10%.2,3,6 This higher rate may reflect the fact that not all of the patients in our study who “discontinued” therapy did so because of adverse effects. Some may have coincidentally stopped treatment for disease progression or treatment futility at approximately the same time that an adverse effect occurred.

The significant difference in survival outcomes seen between patients who discontinue therapy after an IRAE versus those who continue therapy is an intriguing finding with two possible explanations. First, patients have discontinued therapy because of declining health and functional status. In such a situation, prescribers may be making a decision to focus on patient quality of life rather than trying to marginally extend OS. The other possibility is that patients with an IRAE who can continue therapy have increased treatment efficacy along with toxicity, as has been suggested by previous data showing better OS outcomes in patients with IRAE.13

The fact that IM during PD-1 inhibitor treatment does not produce worse treatment outcomes supports a growing body of literature that suggests the same. Two pooled analyses of clinical trial data from metastatic melanoma patients have suggested that response rates are not affected by corticosteroid use during treatment, and an additional retrospective study of 298 metastatic melanoma patients suggests that OS is also not affected.11,15,16 Thus, this study provides data from a larger sample to support the hypothesis that the same patterns seen in melanoma patients may be similar in NSCLC.

In contrast, data from this study suggest that the use of corticosteroids before the initiation of PD-1 therapy may be correlated with worse survival outcomes. In general, patients in clinical trials were required to be free from corticosteroids, with few exceptions, in the period immediately before the initiation of therapy for a period of 3 to 14 days.2-6 Beyond the period of time immediately before the initiation of therapy, the numbers of patients exposed to corticosteroids were not reported. Our data suggest that, either due to interference with the therapeutic effect of PD-1 inhibitors or identification of higher risk patients, this factor may help to identify patients who may have shorter expected survival when treated with PD-1 inhibitors. This effect also seems relatively stable regardless of the time frame used up to 6 months prior to first exposure or the cumulative dose of corticosteroid used. Similar results have been observed in a recent study by Arbour et al17.

As a retrospective study based on database variables, there are a number of potential limitations to this study. One limitation is that our methodology for identifying IRAEs relies on ICD diagnosis codes. Although this allowed us to analyze a large amount of data, it is also subject to the subjective evaluation of different healthcare professionals. Additionally, as a study within Kaiser Permanente, our results are not necessarily generalizable to outside healthcare systems that may not have the level of integration present at Kaiser Permanente. Finally, there is the potential for selection bias because treatment decisions were made based upon clinician judgement rather than on randomization. Regardless, this study provides intriguing data suggesting that the use of corticosteroids may influence or predict survival outcomes when used outside the realm of IRAE management.

In summary, the use of corticosteroids before treatment initiation may be associated with treatment outcomes, whereas the use of IM during treatment may not. This may suggest that, although the use of IM to suppress IRAEs should be performed routinely, the use of corticosteroids before treatment may be more problematic. With further prospective research, we may be able to identify specific criteria regarding the dose, timing, and indication of corticosteroid use to separate out patients that are better suited to immunotherapy with PD-1 inhibitors.

Disclosure Statement

No outside funding supported this study. The authors of this study have no actual or potential conflicts of interest to report, and no funding was required in the preparation of this manuscript. This manuscript had not been accepted for publication or published previously and is not under consideration by any other publication. At the time of this study, Chris Yamamoto was a PGY2 Drug Information Resident with Kaiser Permanente Drug Information Services – California Regions.

Author Affiliations

1Drug Intelligence & Strategy, Kaiser Permanente California Regions, Oakland

2Department of Pharmaceutical Services, Ronald Reagen Medical Center, University of California, Los Angeles

3Pharmacy Outcomes Research Group, Kaiser Permanente California Regions, Downey

4Hematology-Oncology, Kaiser Permanente Southern California Region

5Hematology-Oncology, Kaiser Permanente Southern California Region

6Pharmacy Outcomes Research Group, Kaiser Permanente California Regions, Oakland

Corresponding Author

Timothy Chiu, PharmD, BCPS (timothy.j.chiu@kp.org)

Author Contributions

Timothy Chiu, PharmD, BCPS; Christopher Yamamoto, PharmD, BCPS, BCPS; Fang Niu, MS; and Rita Hui, PharmD, MS, participated in study conception and design, with assistance from Helen Moon, MD; Thach-Giao Truong, MD; and Robert Cooper, MD. Rita Hui, PharmD, MS, and Fang Niu, MS, participated in data collection. Christopher Yamamoto, PharmD, BCPS, BCPS; Rita Hui, PharmD, MS; Fang Niu, MS; Timothy Chiu, PharmD, BCPS; Helen Moon, MD; Thach-Giao Truong, MD; and Robert Cooper, MD, interpreted the data. Christopher Yamamoto, PharmD, BCPS, BCPS; Timothy Chiu, PharmD, BCPS; and Rita Hui, PharmD, MS wrote the manuscript. Helen Moon, MD; Thach-Giao Truong, MD; and Robert Cooper, MD, critically reviewed the manuscript. All authors performed the final approval of the version to be published.

How to Cite this Article

Chiu T, Yamamoto C, Niu F, et al. Immune-related adverse effects associated with programmed death-1 inhibitor therapy in the treatment of non-small cell lung cancer: Incidence, management, and effect on outcomes. Perm J 2020;25:20.034. DOI: 10.7812/TPP/20.034

References

1. Pardoll, DM. The blockade of immune checkpoints in cancer immunotherapy. Nat Rev Cancer 2012 Mar;12(4):252-64. DOI: https://doi.org/10.1038/nrc3239, PMID:22437870

2. Brahmer J, Reckamp KL, Baas P, et al. Nivolumab versus docetaxel in advanced squamous-cell non-small-cell lung cancer. N Engl J Med 2015 Jul;373 (2):123-35. DOI: https://doi.org/10.1056/NEJMoa1504627, PMID:26028407

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 Oct;373 (17):1627-39. DOI: https://doi.org/10.1056/NEJMoa1507643, PMID:26412456

4. Garon EB, Rizvi NA, Hui R, et al Pembrolizumab for the treatment of non-small-cell lung cancer. N Engl J Med 2015 May;372 (21):2018-28. DOI: https://doi.org/10.1056/NEJMoa1501824, PMID:25891174

5. Herbst RS, Baas P, Kim DW, et al Pembrolizumab versus docetaxel for previously treated, PD-L1-positive, advanced non-small-cell lung cancer (KEYNOTE-010): a randomized controlled trial. Lancet 2016 Apr;387 (10027):1540-50. DOI: https://doi.org/10.1016/S0140-6736(15)01281-7, PMID:26712084

6. Reck M, Rodríguez-Abreu D, Robinson AG, et al Pembrolizumab versus chrmotherapy for PD-L1-positive non-small-cell lung cancer. N Engl J Med 2016 Nov;375(19):1823-33. DOI: https://doi.org/10.1056/NEJMoa1606774, PMID:27718847

7. Fetcher LA, Agarwala SS, Hodi FS, et al Ipilimumab and its toxicities: a multidisciplinary approach. Oncologist 2013 Jun;18:733-43. DOI: https://doi.org/10.1634/theoncologist.2012-0483

8. Postow MA, Chesney J, Pavlick AC, et al Nivolumab and ipilimumab versus ipilimumab in untreated melanoma. N Engl J Med. 2015 May;372(21):2006-17. DOI: https://doi.org/10.1056/NEJMoa1414428, PMID:25891304

9. Keytruda [package insert]. Whitehouse Station, NJ: Merck Sharp & Dohme Corp.; 2016.

10. Opdivo [package insert]. Princeton, NJ: Bristol-Myers Squibb Company; 2017.

11. Horvat TZ, Adel NG, Dang TO, et al Immune-related adverse events, need for systemic immunosuppression, and effects on survival and time to treatment failure in patients with melanoma treated with ipilimumab at Memorial Sloan Kettering Cancer Center. J Clin Oncol 2015 Oct;33(28):3193-8. DOI: https://doi.org/10.1200/JCO.2015.60.8448, PMID:26282644

12. Ascierto PA, Simeone E, Sileni VC, et al Clinical experience with ipilimumab 3 mg/kg: real-world efficacy and safety data from an expanded access programme cohort. J Transl Med 2014 May;12:116. DOI: https://doi.org/10.1186/1479-5876-12-116, PMID:24885479

13. Freeman-Keller M, Kim Y, Cronin H, et al Nivolumab in resected and unresected metastatic melanoma: characteristics of immune-related adverse events and association with outcomes. Clin Cancer Res 2016 Feb;22(4):886-94. DOI: https://doi.org/10.1158/1078-0432.CCR-15-1136, PMID:26446948

14. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987 Jan;40(5):373-83. DOI: https://doi.org/10.1016/0021-9681(87)90171-8, PMID:3558716

15. Downey SG, Klapper JA, Smith FO, et al Prognostic factors related to clinical response in patients with metastatic melanoma treated by CTL-associated antigen-4 blockade. Clin Cancer Res 2007 Nov;13(22 Pt 1):6681-8. DOI: https://doi.org/10.1158/1078-0432.CCR-07-0187, PMID:17982122

16. Weber JS, Hodi FS, Wolchok JD, et al Safety profile of nivolumab monotherapy: a pooled analysis of patients with advanced melanoma. J Clin Oncol 2017 Mar;35(7):785-92. DOI: https://doi.org/10.1200/JCO.2015.66.1389, PMID:28068177

17. Arbour KC, Mezquita L, Long N et al Impact of baseline steroids on efficacy of programmed cell death-1 and programmed death-ligand 1 blockade in patients with non-small-cell lung cancer. J Clin Oncol 2018 Oct;36(28):2872. DOI: https://doi.org/10.1200/JCO.2018.79.0006, PMID:30125216

18. Eggermont AMM, Kicinski M, Blank CU et al Association between immune-related adverse events and recurrence-free survival among patients with stage III melanoma randomized to receive pembrolizumab or placebo: a secondary analysis of a randomized clinical trial. JAMA Oncol 2020 Apr;6(4):519-27. DOI: https://doi.org/10.1001/jamaoncol.2019.5570.

Keywords: cancer, lung cancer, non-small cell lung cancer, checkpoint inhibitor, nivolumab, pembrolizumab, IRAE, immune-related adverse event, corticosteroids, immunosuppression

ETOC

Click here to join the eTOC list or text ETOC to 22828. You will receive an email notice with the Table of Contents of The Permanente Journal.

CIRCULATION

2 million page views of TPJ articles in PubMed from a broad international readership.

Indexing

Indexed in MEDLINE, PubMed Central, EMBASE, EBSCO Academic Search Complete, and CrossRef.


                                             

 

 

ISSN 1552-5775 Copyright © 2021 thepermanentejournal.org

All Rights Reserved