Abstract
Objectives
Immune checkpoint inhibitor immunotherapy (IO) is revolutionizing cancer care but can lead to significant toxicity. This study seeks to describe potential risk factors for immune-related adverse events (irAEs) specifically among older adults.
Materials and Methods
This was a retrospective study at a single academic comprehensive cancer center based on chart review data abstracted by physicians. For patients aged ≥70 years, frequency, type, and grade of irAEs and their association with baseline patient demographics, comorbidities, mobility, and functional status were characterized using bivariate analysis. Based on those results, multivariable logistic regressions were constructed to model the association between these characteristics with any grade and grade 3 or higher irAEs.
Results
Data were analyzed for 238 patients aged ≥70 years who received IO for mostly (≥90%) advanced cancer between 2011 and 2018. Thirty-nine percent of older adults experienced an irAE and 13% experienced one that was grade 3 or higher. In the multivariable analysis, depression was associated with an increased incidence of any grade irAE, while decreased life-space mobility was associated with an increased incidence of grade ≥3 irAEs.
Conclusion
Most characteristics of special interest among older adults, include fall risk, weight loss, cognitive limitations, and hearing loss, were not associated with irAEs in our study. However, decreased life-space mobility and depression are potential risk factors for IO toxicity among older adults with advanced cancer. Interventions designed to evaluate and mitigate modifiable risk factors for treatment-related toxicity are needed, and the results of this study may be useful for guiding those efforts.
Keywords: checkpoint inhibitor, immunotherapy, toxicity, geriatrics, immune-related adverse events
Immune checkpoint inhibitor immunotherapy is revolutionizing cancer care but can lead to significant toxicity. This article describes potential risk factors for immune-related adverse events among older adults.
Implications for Practice.
Most characteristics of special interest among older adults, including fall risk, weight loss, cognitive limitations, and hearing loss, do not appear to be significantly associated with immune-related adverse events, supporting the safety of immune checkpoint inhibitor immunotherapy in the older adult population. However, decreased mobility and depression are potential risk factors for immunotherapy toxicity among older adults with advanced cancer. Interventions designed to evaluate and mitigate modifiable risk factors for treatment-related toxicity are needed
Introduction
Over the past 10 years, immune checkpoint inhibitor immunotherapy (IO) has become a standard of care treatment for several cancers that commonly affect older adults, particularly lung cancers, melanoma, and genitourinary cancers.1 There is a plethora of evidence that IO produces improved cancer response and survival rates among the general population.2 However, the clinical trials from which data regarding efficacy and toxicity of these drugs were obtained included primarily younger patients (age <70 years) with good performance status, and they did not report safety and efficacy data by either chronologic or biologic age.1,3 Consequently, despite the fact that these drugs are commonly used among older adults, who represent over 50% of all new cancer diagnoses, data on IO treatment outcomes among older adults are improving but still limited.1-5
Cancer propagation and progression require tumor cells to escape the immune system via complex mechanisms of downregulation and inhibition of immune cell functions. Checkpoint inhibitors are monoclonal antibodies that inhibit the downregulation of the immune system by cancer cells.6-9 Importantly, these therapies augment the ability of the cell-mediated immune system, which plays a key role in tumor defense, to recognize tumor cells, and inhibit their survival.6,7 The 2 main classes of IO drugs currently in use are cytotoxic T-lymphocyte antigen-4 (CTLA-4) inhibitors and inhibitors of the programmed death protein/ligand-1 (PD-1/L1) system. These work by overriding T-cell inhibition, resulting in a bolstered immune response against cancer cells.10 This unique mechanism of action also results in the impairment of physiologic immune self-tolerance, which normally prevents the stimulation of an immune response against normal human tissues.10 As a result, IO can cause autoimmune-like conditions known as immune-related adverse events (irAEs) that are distinct from side effects such as nausea, hair loss, mucositis, and neuropathy, commonly seen with traditional cytotoxic chemotherapy.8,10,11
Common irAEs that have been described during and after treatment with IO include primarily auto-inflammatory conditions such as dermatitis, hyper and hypothyroidism, pneumonitis, colitis, hepatitis, type 1 diabetes, myocarditis, hypopituitarism, and hemolytic anemia.8,12 These adverse events are common with IO use; for example, Lichtenstein et al reported irAEs in 42% of patients receiving IO for non-small cell lung cancer (NSCLC).13 The risk for toxicity increases dramatically, up to 3-fold, when IO agents are used in combination with each other, with some studies reporting irAEs in more than 90% of patients receiving dual checkpoint-inhibitor therapy. In these cases, over half of the irAEs were grade 3 or higher, resulting in hospitalization rates of up to 36%.14-16 Treatment for irAEs usually involves immune suppressive therapies such as corticosteroids, which are associated with risk for steroid-induced psychosis, especially in older patients.16,17
As the number of older adults diagnosed with and treated for cancer grows, the importance of using high-quality geriatric assessments (GAs) to guide oncologists’ treatment of older adult patients is becoming increasingly recognized. While many studies have found the incidence of irAEs to be comparable between younger and older adults, data on the risk factors for toxicity in older adults are notably lacking.13,15 Recent studies have shown that chronological age or performance status alone has limited value in predicting risk of IO toxicity, further highlighting the need to consider age-independent geriatric variables.18 Additionally, a growing understanding of immunosenescence contends that aging produces physiologic changes in not only immune function but also composition, with studies showing pro-inflammatory cells to be predominant later in life.19,20 These immune changes could contribute to producing a different IO response and toxicity profile in older adults compared to younger, immuno-typical patients. In the general population, the most noted risk factors for irAEs are a personal or family history of autoimmune conditions, past viral infections, and medications with potential autoimmune toxicities.21,22 However, the association of irAEs with aging and independent geriatric-specific characteristics is still poorly established.5
Functional impairment, malnutrition, depressive symptoms, and medical comorbidities have been shown to be independent predictors of severe chemotherapy toxicity among older adults with cancer, and these have been incorporated into well-validated instruments for predicting chemotherapy toxicity.23-29 While this utility is well-established for chemotherapy, there is little data regarding the role of a geriatric assessment (GA),18,23 in predicting outcomes and toxicity among older adult patients receiving IO, a population shown to have a high prevalence of impairments within GA domains.30 Thus, our study aims to describe the clinical and functional, geriatrics-focused characteristics of a large cohort of older adults with advanced cancer receiving IO, and to determine possible associations between these characteristics and irAEs.
Methods
Study Design and Sample
This is a single-institution, retrospective cohort study. It was approved by the institutional review board at The Ohio State University Comprehensive Cancer Center (OSUCCC). Older adult patients (age ≥70 years) with any stage cancer who received at least one dose of IO between October 6, 2011, and April 5, 2018, at the OSUCCC were included. Patients receiving IO in standard of care, clinical trial, and off-label settings were included in the study; there were no exclusion criteria. Pharmacy administration dates were used to validate the receipt of IO.
Construction of Variables
Patients’ demographic and clinical information (including sex, race, body mass index, cancer type and stage based on pathology and imaging reports, dates and types of treatments received, date of death or last known alive, and ECOG performance status) was abstracted by physicians from the electronic medical record and entered into a REDCap database (Vanderbilt University, v8.10.10).31 Baseline functional characteristics (including mobility, ADL/IADL limitations, and cognitive limitations) of patients within 30 days of IO start were abstracted using clinician, nursing, and physical and occupational therapy notes (see Supplementary Table S1 for variable definitions). Specifically, subjective history, physical exam, and physical therapy documentation were used for abstracting functional and mobility status, while nursing flowsheets were reviewed for information regarding assistive device use and weight changes. Importantly, the concept of mobility in this study refers to life-space mobility and is an inference of information that would traditionally be gathered using a survey tool such as the Life-Space Assessment.32 In contrast, physical mobility disability is inferred through information regarding the use of an assistive device such as a cane, walker, or wheelchair (Supplementary Table S1). If no information could be found to unambiguously support the presence or absence of a variable for a specific patient, then this data point was considered “unavailable” for statistical analysis purposes.
Detailed information regarding irAEs, including date of diagnosis, attribution, and clinical interventions, was abstracted. IrAEs were graded using the Common Terminology Criteria for Adverse Events, version 4.0. IrAEs were defined by the treating oncologist or multi-disciplinary team based on the following criteria as per prior studies: (1) timing of irAE occurring after initiation of IO; (2) pathologic diagnosis, when available; (3) exclusion of alternate etiologies; or (4) clinical improvement with appropriate irAE-directed therapies.33,34 For any patients receiving concurrent medications with IO, such as chemotherapy, attribution of irAEs to IO as opposed to the concurrent medications was determined by review of physician documentation.
Outcomes
The primary outcome was identification of patient characteristics including demographics, comorbidities, mobility, and functional status associated with any grade and grade 3 or higher irAEs. Secondary outcomes included the incidence, severity, and type of irAEs.
Statistical Analysis
First, baseline demographic, clinical, and functional patient characteristics were summarized using descriptive statistics including medians and interquartile range (IQR) for the continuous variables and frequencies for the categorical variables. Next, for the bivariate analysis, chi-squared tests and Wilcoxon rank-sum tests were used to determine patient characteristics associated with any grade and grade 3 or higher IO toxicity. Multivariable logistic regressions were used to develop models of characteristics associated with any grade and grade 3 or higher IO toxicity. Variables with a P < .05 in the bivariate analysis were entered into a multivariable model. A P-value of < .05 was considered statistically significant. All analyses were conducted in SAS version 9.4 (SAS Institute, Cary, NC). For the bivariate analysis procedure, patients with missing data for any variable were excluded from the bivariate analysis for that variable. However, data for all variables, regardless of missingness, were reported in this manuscript for descriptive purposes. Characteristics within the bivariate analysis with missing data from a large portion of patients (20% or greater) were excluded from consideration for the multivariable analysis. For each model within the multivariable analysis, patients with an unavailable data point for any variable in that model were excluded from the patient sample used in the model analysis.
Results
Baseline Characteristics
In total, we identified 238 patients who fit our inclusion criteria (Table 1). The cohort was 59% male and 95% White/Caucasian. The median age was 76 years, with an IQR of 73-80 years. The most common cancer type was melanoma (n = 85, 36%), followed by NSCLC (n = 56, 24%). Two hundred and fifteen (90.3%) were confirmed to have stage III or IV disease. Nivolumab was the most used IO agent (50%), followed by pembrolizumab (19%) and ipilimumab (18%). Only one patient received concurrent chemotherapy with IO. Regarding medical history, 77% of patients had 1-4 comorbid conditions, 65% had a body mass index (BMI) >25, and 65% were current or former smokers. At baseline, 24% had an Eastern Cooperative Oncology Group (ECOG)35 performance status ≥2. Median duration of IO therapy was 77.5 days (IQR 42-203 days); 40% of patients received IO as a first-line agent for cancer treatment and 29% for third-line and beyond.
Table 1.
Baseline characteristics of all patients.
| Characteristic | Frequency, n (%) (n = 238) |
|---|---|
| Age, years | 76 [73-80]a |
| 70-74 | 93 (39) |
| 75-79 | 76 (32) |
| ≥80 | 69 (29) |
| Gender | |
| Male | 140 (59) |
| Female | 97 (41) |
| Not documented | 1 (0) |
| Race | |
| White | 225 (95) |
| Non-White | 13 (6) |
| BMI | 27.3 [24-90.3]a |
| Underweight (<18) | 5 (2) |
| Normal (18-25) | 75 (32) |
| Overweight (25.1-30) | 91 (38) |
| Obese (>30) | 65 (27) |
| Not documented | 2 (1) |
| History of smoking | 154 (65) |
| Number of comorbidities | 3 (2-4)a |
| 0 | 12 (5) |
| 1-2 | 89 (37) |
| 3-4 | 96 (40) |
| ≥5 | 41 (17) |
| Cancer type | |
| Melanoma | 85 (36) |
| Non-small cell lung cancer | 56 (24) |
| Renal cell carcinoma | 20 (8) |
| Other | 77 (17) |
| Cancer stage | |
| II | 2 (1) |
| III | 17 (7) |
| IV | 198 (83) |
| Other or not specified | 21 (9) |
| IO agent used | |
| Atezolizumab | 10 (4) |
| Ipilimumab | 44 (18) |
| Nivolumab | 119 (50) |
| Ipilimumab + nivolumab | 11 (5) |
| Pembrolizumab | 46 (19) |
| Other | 8 (3) |
| Prior therapies | |
| Any radiation therapy | 118 (50) |
| Brain radiation | 30 (13) |
| Chemotherapy | 117 (49) |
| ECOG performance status | |
| 0 | 78 (33) |
| 1 | 102 (43) |
| 2 | 51 (21) |
| 3 | 6 (3) |
| Not documented | 1 (0) |
| Line of therapy | |
| 1st | 97 (41) |
| 2nd | 70 (29) |
| 3rd or beyond | 69 (29) |
| Not documented | 2 (1) |
| IO duration, weeks | |
| ≤6 | 64 (26.9) |
| 6-12 | 60 (25.2) |
| 12-24 | 42 (17.7) |
| ≥24 | 72 (30.3) |
aMedian [interquartile range].
Abbreviations: BMI, body mass index (kg/m2); ECOG, Eastern Cooperative Oncology Group; IO, immunotherapy.
Geriatrics-Focused Characteristics
The most prevalent geriatrics-focused characteristics were weight loss or nutrition risk (48% of overall cohort, detailed in Table 2), fall history or fall risk (42%), physical mobility disability/assistive device use (39%), and hearing impairment (31%). Of note, data on independent activities of daily living (IADL) limitations were missing for 50% of patients and on daily living (ADL) limitations for 20% of patients.
Table 2.
Bivariate analysis of baseline patient characteristics and their association with irAEs. Characteristics with significant associations (P < .05) and that did not have missing data points for more than 20% of patients were used for the multivariable analyses.
| Characteristic | Frequency, no. (%) | Characteristic not documented no. (%) Total n = 238 |
P-value (Χ2) any grade irAE |
P-value (Χ2) Grade ≥3 irAEs |
||
|---|---|---|---|---|---|---|
| No irAEs n = 146 |
Any grade irAE n = 92 |
Grade ≥3 irAEs n = 32 |
||||
| Age, years+ | 76 [73-80] | 76 [73-80.5] | 74.5 [73-81] | 0 | .45 | .53 |
| 70-74 | 54 (37) | 39 (42) | 16 (50) | |||
| 75-79 | 51 (35) | 25 (27) | 7 (22) | |||
| ≥ 80 | 46 (32) | 28 (30) | 9 (28) | |||
| Female sex | 61 (42) | 36 (39) | 15 (47) | 1 (0) | .66 | .27 |
| BMI+ | 27.2 [24-30.3] | 27.8 [24-30.7] | 26.9 [25-29.8] | 2 (1) | .35 | .54 |
| Underweight (<18) | 5 (3) | 0 (0) | 5 (2) | |||
| Normal (18-25) | 45 (31) | 30 (33) | 9 (28) | |||
| Overweight (25.1-30) | 65 (45) | 26 (28) | 8 (25) | |||
| Obese (>30) | 29 (20) | 36 (39) | 15 (27) | |||
| Number of comorbidities+ | 3 [2-4] | 3 [2-4] | 3 [2-4] | 0 | .58 | .74 |
| 0 | 6 (4) | 6 (7) | 1 (3) | |||
| 1-2 | 56 (38) | 33 (36) | 13 (41) | |||
| 3-4 | 56 (38) | 40 (43) | 14 (44) | |||
| 5+ | 28 (19) | 13 (14) | 4 (13) | |||
| Cancer type | 0 | .08 | .75 | |||
| Melanoma | 46 (32) | 39 (42) | 13 (41) | |||
| NSCLC | 35 (24) | 21 (23) | 9 (28) | |||
| Renal cell | 10 (7) | 10 (11) | 5 (16) | |||
| Other | 55 (38) | 22 (24) | 5 (16) | |||
| Prior therapies | 5 (2) | |||||
| Chemotherapy | 81 (55) | 36 (39) | 2 (6) | .01 | .51 | |
| Radiation therapy | 76 (52) | 42 (46) | 13 (41) | .27 | .56 | |
| ECOG performance status | 1 (0) | .71 | .37 | |||
| 0 | 78 (53) | 0 | 11 (34) | |||
| 1 | 68 (47) | 34 (37) | 10 (31) | |||
| 2 or higher | 1 (1) | 56 (61) | 11 (34) | |||
| Line of therapy | 2 (1) | <.01 | .07 | |||
| First | 49 (34) | 48 (52) | 15 (47) | |||
| Second | 44 (30) | 26 (28) | 7 (22) | |||
| Third or beyond | 52 (36) | 17 (18) | 10 (31) | |||
| Clinical response to IO | 31 (21) | 27 (29) | 25 (78) | 0 | .16 | .25 |
| History of smoking | 100 (68) | 54 (59) | 17 (53) | 0 | .12 | .43 |
| Geriatric variables | ||||||
| Depression | 26 (18) | 29 (32) | 10 (31) | 0 | .01 | .97 |
| Daily medications ≥ 10 | 25 (17) | 18 (20) | 9 (28) | 1 (0) | .53 | .53 |
| Physical mobility disability | 58 (40) | 35 (38) | 15 (47) | 27 (11) | .93 | .44 |
| Hearing impairment | 50 (34) | 24 (26) | 9 (28) | 1 (0) | .26 | .78 |
| Cognitive limitations | 17 (12) | 12 (13) | 5 (16) | 1 (0) | .42 | .33 |
| ADL limitations | 25 (17) | 10 (11) | 3 (9) | 48 (20) | .30 | .07 |
| IADL limitations | 29 (20) | 15 (16) | 10 (31) | 118 (50) | .55 | <.01 |
| Weight loss/nutrition risk | 77 (53) | 38 (41) | 17 (53) | 0 | .09 | .09 |
| Fall history or risk | 61 (42) | 39 (42) | 15 (47) | 2 (1) | .94 | .29 |
| Impaired life-space mobility | 28 (19) | 17 (18) | 11 (34) | 30 (13) | .55 | <.01 |
+Median [interquartile range].
Abbreviations: ADL, activities of daily living; BMI: body mass index (kg/m2); ECOG, Eastern Cooperative Oncology Group; IADL, independent activities of daily living; IO, immunotherapy; NSCLC, non-small cell lung cancer.
Toxicity
Any-grade irAEs were seen in 39% of patients, with dermatitis being most common (16%), followed by colitis or diarrhea (8%), and abnormal thyroid function (8%). Grade 3 or higher irAEs were seen in 13% of patients and included colitis or diarrhea (4%), pneumonitis (3%), hepatitis (3%), and dermatitis (3%).
Bivariate Analysis
Bivariate analysis of baseline and geriatrics-focused characteristics with IO toxicity is shown in Table 2. Among baseline patient characteristics, history of prior chemotherapy (P = .01) and line of therapy (P < .01) were associated with any-grade irAE. For the geriatrics-focused covariates, depression was associated with any-grade irAE (P = .01), while IADL limitations (P < .01) and impaired life-space mobility (P < .01) were associated with high-grade irAEs. Depression was not significantly associated with impaired life-space mobility (P = .21).
Multivariable Analysis
Baseline and geriatrics-focused characteristics that were shown to be significantly associated with IO toxicity (P < .05) and did not have missing data points for more than 20% of patients were tested for significance with multivariable analyses (Table 3). In the final model that included age group and line of therapy, depression was associated with any-grade toxicity (OR = 2.41 [1.24-4.71], P < .01), while third-line IO therapy was associated with decreased risk of any-grade toxicity when compared to first-line therapy (OR = 0.30 [0.15-4.71], P < .01). In the final model that included age group, impaired life-space mobility was associated with grade 3 or higher toxicity (OR = 4.4 [1.4-4.5], P = .01). Age group was not significantly associated with any-grade or high-grade toxicity.
Table 3.
Multivariable analyses of potential risk factors for any grade and grade 3 or higher toxicity. In these final models, depression showed a significant association with any grade toxicity while impaired mobility showed a significant association with grade 3 or higher toxicity.
| Covariate | Odds ratio (95% CI) | P-value | |
|---|---|---|---|
| Any grade toxicity | Age, years | ||
| 70-74 | 0.94 (0.47-1.86) | .71 | |
| 75-79 | 0.70 (0.35-1.43) | .30 | |
| ≥80 | Reference | — | |
| Depression | |||
| Yes | 2.41 (1.24-4.71) | <.01 | |
| No | Reference | — | |
| IO line of therapy | |||
| 3rd or higher | 0.30 (0.15-0.61) | <.01 | |
| 2nd | 0.63 (0.33-1.20) | .65 | |
| 1st | Reference | — | |
| Grade ≥3 toxicity | Age, years | ||
| 70-74 | 1.74 (0.56-5.42) | .64 | |
| 75-79 | 1.42 (0.38-5.37) | .90 | |
| ≥80 | Reference | — | |
| Life-space mobility | |||
| Impaired | 4.4 (1.4-14.5) | .01 | |
| Full | Reference | — | |
Abbreviation: IO, immunotherapy.
Discussion
Among a cohort of older adults ≥70 years with mostly (≥90%) advanced cancer, this study found a significant association between medical treatment for depression and any-grade irAEs, as well as impaired life-space mobility with grade 3 or higher irAEs. The prevalence of irAEs in this older adult cohort (39% and 13% for any-grade and high-grade irAEs) is similar to retrospective data previously reported for adults age <70 years at our institution (28.7% and 14.5% for any-grade and high-grade irAEs, respectively).36
Our findings are consistent with existing literature on the association of chemotherapy toxicity with depression or negative mood,37,38 and more recent data among adults with advanced non-small cell lung cancer.39,40 However, the mechanisms of these associations remain understudied. One possible explanation for increased rates of depression is the symptom perception hypothesis, which proposes that negative emotions can create a psychosomatic disposition to adverse physiologic reactions.41 This hypothesis has been cited as a rationale for the increased frequency of physical symptom reporting observed in patients with depression,42 and it is possible that patients in our study with underlying depression were more likely to report lower-grade toxicities such as mild diarrhea or dermatitis that would have otherwise been undetected on clinical evaluation. Importantly, specific data regarding depression symptoms or severity were not available. In considering biochemical mechanisms, studies have shown that emotional stressors, including depression are associated with the upregulation of stress hormones and subsequent activation of pro-inflammatory pathways.43,44 Since IO efficacy and toxicity both stem from an auto-inflammatory response, it is reasonable that the increased inflammatory burden associated with depression could increase the risk for inflammation-mediated irAEs. This association indicates that screening for depression and ensuring that depression is clinically well managed may be important for IO toxicity outcomes.
Our results also showed a significant relationship between decreased mobility and risk of high-grade irAEs. However, our definition of impaired life-space mobility by the amount of time spent at home rather than out in the community did not separate medical or physical versus psychosocial factors, such as limited access to transportation, as contributing to mobility impairment. Importantly, we did not find a significant association between medical treatment for depression and decreased life-space mobility. Future studies to understand the underlying causes of life-space mobility impairment are needed for a better understanding of their associations with irAEs.
Regardless of mechanism, the association between certain geriatric-focused clinical variables and irAEs suggests that an evaluation of these variables in older patients may be helpful for identifying those at increased risk for irAEs, which can be severe and significantly impair quality of life. The importance and benefits of using a GA to evaluate fitness and other measurable characteristics of older adults prior to initiation of intensive cancer treatment have been well described and shown to be effective in predicting morbidity and mortality.23 A GA should include an evaluation of a patient’s functional status and fall risk, comorbid medical conditions, cognition, psychological state, social support, nutrition status, and medications.3,23 When these components are incorporated, GAs have shown efficacy in identifying areas of vulnerabilities and potential intervention that may otherwise be overlooked in a usual cancer care clinical evaluation.3,25 More recently, prospective randomized clinical trials, GAP70+ and GAIN, have demonstrated reduced toxicity of cytotoxic chemotherapy among older adult patients assigned to undergo GA and, in the case of the GAIN study, interventions based on assessment recommendations.45,46
Given the limited number of GA measures that were found to be associated with irAE risk, our findings need to be validated in a prospective manner before more concrete conclusions can be drawn regarding the utility of the GA for informing IO treatment risks in older adults. Additionally, our study found that many geriatrics-focused variables, including cognitive limitations, hearing impairment, and assistive mobility device use (such as walker, cane, or wheelchair), did not appear to be associated with increased risk for irAEs. While promising, these findings require further and more comprehensive evaluation through the use of a formal geriatric assessment and analysis of patient-reported data.
Limitations
Our study has notable limitations. As a single-center, retrospective study, the generalizability of our findings is limited, and no causality can be determined. All variables within our study were assessed via chart review, with patients of various providers and multidisciplinary teams within our institution being included (Supplementary Table S1). Additionally, some of the variables (eg, cognitive limitations) analyzed in this retrospective study were not described using objective measures. As a result, the presence or absence of a variable was subject to differences in patient and/or family reporting, and provider interpretation and documentation. Finally, some variables, most notably IADL limitations, were inconsistently assessed and documented by providers. This resulted in missing data points for a subset of patients, which were excluded from the multivariable analyses. On a similar note, at least some of the geriatrics-focused characteristics were likely underreported in clinician documentation; for example, some patients may have inaccurately been documented as not having cognitive deficits. For all of these reasons, validation of the characteristics described in our study in a prospective cohort is needed. Validated methods for collecting data on these characteristics, such as natural language processing of medical record data and patient subjective activity reporting, are also needed.
Conclusion
Our study offers a novel contribution to the collective understanding of IO toxicity and its risk factors among older adults with advanced cancer. We found significant associations between impaired life-space mobility with high-grade toxicity and depression with any-grade toxicity. Overall, we found that most characteristics of special interest among older adults are not associated with IO toxicity, but further work is needed to confirm the overall safety and use of IO in this population. Given the small scale of our cohort and the retrospective nature of this analysis, which is likely limited by underreporting in clinical documentation, this study is primarily hypothesis-generating. These results can inform future studies with larger, more diverse patient groups and more comprehensive methods of measuring geriatrics-focused characteristics to strengthen the generalizability of our findings. Treatment efficacy with respect to geriatrics-focused patient characteristics should be assessed in addition to toxicity. Moreover, prospective, intervention-focused studies are necessary to inform guidelines of care to improve outcomes, patient satisfaction, and quality of life among older adults at risk for IO toxicity.
Supplementary Material
Acknowledgments
We wish to acknowledge the contributions of Angela Dahlberg, BSJ, Editor, Division of Medical Oncology, Department of Internal Medicine, The Ohio State University Comprehensive Cancer Center, in the revision of this manuscript.
Contributor Information
Andrew C Johns, Department of Internal Medicine, The Ohio State University Wexner Medical Center, Columbus, OH, USA.
Mike Yang, College of Medicine, The Ohio State University, Columbus, OH, USA.
Lai Wei, Department of Biomedical Informatics, The Ohio State University, Columbus, OH, USA.
Madison Grogan, Division of Medical Oncology, Department of Internal Medicine, The Ohio State University Comprehensive Cancer Center, Columbus, OH, USA.
Daniel Spakowicz, Department of Biomedical Informatics, The Ohio State University, Columbus, OH, USA; Division of Medical Oncology, Department of Internal Medicine, The Ohio State University Comprehensive Cancer Center, Columbus, OH, USA.
Sandipkumar H Patel, Division of Medical Oncology, Department of Internal Medicine, The Ohio State University Comprehensive Cancer Center, Columbus, OH, USA.
Mingjia Li, Division of Hospital Medicine, Department of Internal Medicine, The Ohio State University Wexner Medical Center, Columbus, OH, USA.
Marium Husain, Division of Medical Oncology, Department of Internal Medicine, The Ohio State University Comprehensive Cancer Center, Columbus, OH, USA.
Kari L Kendra, Division of Medical Oncology, Department of Internal Medicine, The Ohio State University Comprehensive Cancer Center, Columbus, OH, USA.
Gregory A Otterson, Division of Medical Oncology, Department of Internal Medicine, The Ohio State University Comprehensive Cancer Center, Columbus, OH, USA.
Ashley E Rosko, Division of Hematology, Department of Internal Medicine, The Ohio State University Wexner Medical Center, Columbus, OH, USA.
Barbara L Andersen, Department of Psychology, The Ohio State University, Columbus, OH, USA.
David P Carbone, Division of Medical Oncology, Department of Internal Medicine, The Ohio State University Comprehensive Cancer Center, Columbus, OH, USA.
Dwight H Owen, Division of Medical Oncology, Department of Internal Medicine, The Ohio State University Comprehensive Cancer Center, Columbus, OH, USA.
Carolyn J Presley, Division of Medical Oncology, Department of Internal Medicine, The Ohio State University Comprehensive Cancer Center, Columbus, OH, USA.
Funding
Research support was provided by the REDCap project and The Ohio State University Center for Clinical and Translational Science grant support (National Center for Advancing Translational Sciences, Grant UL1TR002733). Dr Presley and Dr Owen were Paul Calabresi Scholars supported by the OSU K12 Training Grant for Clinical Faculty Investigators (K12 CA133250). Dr. Presley is currently supported by the National Institute of Aging: 1K76AB074923-01. Research reported in this publication was supported by The Ohio State University Comprehensive Cancer Center and the National Institutes of Health under grant number P30 CA016058. We thank the Biostatistics Shared Resource (BSR) at The Ohio State University Comprehensive Cancer Center, Columbus, OH, for biostatistical support of this study. Institutions that provided funding support had no role in the design or conduct of this study or in preparation of the manuscript.
Conflict of Interest
David P. Carbone reported financial relationship with James Cancer Center, Abbvie, Bayer, Boehringer Ingelheim, Daiichi Sankyo, EMD Serono, GlaxoSmithKline, Incyte, Inivata, Inovio Pharmaceuticals, Janssen, Kyowa Hakko Kirin, Loxo, Merck, Novartis, Pfizer, Takeda, Bristol Myers-Squibb KK (Japan), Flame Biosciences, G1 Therapeutics (Intellisphere), GenePlus, Genentech/Roche, Gloria Biosciences, Lilly, MSD, Novocure, Oncocyte, OncoHost, Piper Sandler, Amgen, Curio Science, Johnson & Johnson, Merck KGaA, Novartis, Sanofi, personal fees and other from AstraZeneca amd Nexus Pharmaceuticals Inc., grants and other from Bristol-Myers Squibb. Kari L. Kendra has received institution-directed research funding from Bristol-Myers Squibb, Merck, GlaxoSmithKline, Novartis, Immunocore, Karyopharm, and Medpace. Dwight H. Owen reported receiving institution-directed research funding from Bristol-Myers Squibb, Merck, Genentech, Palbiofarma, and Onc.AI; he reports consultant work for theMednet.org. The other authors indicated no financial relationships. All listed conflicts of interest are outside the scope of this article.
Author Contributions
Conception/design: A.C.J., C.J.P. Provision of study material or patients: A.C.J., M.G., S.H.P., M.L., M.H., K.L.K., G.A.O., D.P.C., D.H.O. Collection and/or assembly of data: A.C.J., M.G., S.H.P., M.L., M.H., K.L.K., G.A.O., D.P.C., D.H.O. Data analysis and interpretation: A.C.J., L.W., D.S., A.E.R., B.L.A., C.J.P. Manuscript writing: A.C.J., M.Y., B.L.A., D.H.O., C.J.P. Final approval of manuscript: all authors.
Data Availability
The data underlying this article will be shared on reasonable request to the corresponding author.
Previous Presentation
This work was previously presented at the American Geriatrics Society 2019 Annual Scientific Meeting.
References
- 1. Marrone KA, Forde PM.. Cancer immunotherapy in older patients. Cancer J. 2017;23(4):219-222. 10.1097/PPO.0000000000000268. [DOI] [PubMed] [Google Scholar]
- 2. Elias R, Giobbie-Hurder A, McCleary NJ, et al. Efficacy of PD-1 & PD-L1 inhibitors in older adults: a meta-analysis. J ImmunoTher Cancer. 2018;6(1):26. 10.1186/s40425-018-0336-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Loh KP, Soto-Perez-de-Celis E, Hsu T, et al. What every oncologist should know about geriatric assessment for older patients with cancer: Young International Society of Geriatric Oncology Position Paper. J Oncol Pract. Feb 2018;14(2):85-94. 10.1200/JOP.2017.026435. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. NCI. Age and Cancer Risk. NIH. Accessed August 11, 2021. https://www.cancer.gov/about-cancer/causes-prevention/risk/age [Google Scholar]
- 5. Presley CJ, Gomes F, Burd CE, Kanesvaran R, Wong ML.. Immunotherapy in older adults with cancer. J Clin Oncol. 2021;39(19):2115-2127. 10.1200/JCO.21.00138. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Spitzer MH, Carmi Y, Reticker-Flynn NE, et al. Systemic immunity is required for effective cancer immunotherapy. Cell. 2017;168(3):487-502.e15. 10.1016/j.cell.2016.12.022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Ashine TA, Tesema T.. Exploitation of cell mediated immune responses to cancer immunotherapy. J Med, Physiol and Biophy 2019;59:17-29. 10.7176/jmpb/59-03. [DOI] [Google Scholar]
- 8. Johnson DB, Chandra S, Sosman JA.. Immune checkpoint inhibitor toxicity in 2018. JAMA. 2018;320(16):1702-1703. 10.1001/jama.2018.13995. [DOI] [PubMed] [Google Scholar]
- 9. Hargadon KM, Johnson CE, Williams CJ.. Immune checkpoint blockade therapy for cancer: an overview of FDA-approved immune checkpoint inhibitors. Int Immunopharmacol. 2018;62:29-39. 10.1016/j.intimp.2018.06.001. [DOI] [PubMed] [Google Scholar]
- 10. Spain L, Diem S, Larkin J.. Management of toxicities of immune checkpoint inhibitors. Cancer Treat Rev. 2016;44:51-60. 10.1016/j.ctrv.2016.02.001. [DOI] [PubMed] [Google Scholar]
- 11. Hurria A, Togawa K, Mohile SG, et al. Predicting chemotherapy toxicity in older adults with cancer: a prospective multicenter study. J Clin Oncol. 2011;29(25):3457-3465. 10.1200/JCO.2011.34.7625. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Varricchi G, Galdiero MR, Marone G, et al. Cardiotoxicity of immune checkpoint inhibitors. ESMO Open. 2(4):e000247e000247. 10.1136/esmoopen-2017-000247. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Lichtenstein MRL, Nipp RD, Muzikansky A, et al. Impact of age on outcomes with immunotherapy in patients with non-small cell lung cancer. J Thorac Oncol. 2019;14(3):547-552. 10.1016/j.jtho.2018.11.011. [DOI] [PubMed] [Google Scholar]
- 14. Shoushtari AN, Friedman CF, Navid-Azarbaijani P, et al. Measuring toxic effects and time to treatment failure for nivolumab plus ipilimumab in melanoma. JAMA Oncol. 2018;4(1):98-101. 10.1001/jamaoncol.2017.2391. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Friedman CF, Wolchok JD.. Checkpoint inhibition and melanoma: considerations in treating the older adult. J Geriatr Oncol. 2017;8(4):237-241. 10.1016/j.jgo.2017.04.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Boutros C, Tarhini A, Routier E, et al. Safety profiles of anti-CTLA-4 and anti-PD-1 antibodies alone and in combination. Nat Rev Clin Oncol. 2016;13(8):473-486. 10.1038/nrclinonc.2016.58. [DOI] [PubMed] [Google Scholar]
- 17. Dubovsky AN, Arvikar S, Stern TA, Axelrod L.. The neuropsychiatric complications of glucocorticoid use: steroid psychosis revisited. Psychosomatics. 2012;53(2):103-115. 10.1016/j.psym.2011.12.007. [DOI] [PubMed] [Google Scholar]
- 18. Gomes F, Lorigan P, Woolley S, et al. A prospective cohort study on the safety of checkpoint inhibitors in older cancer patients—the ELDERS study. ESMO Open. 2021;6(1):100042. 10.1016/j.esmoop.2020.100042. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Hurez V, Padron A, Svatek RS, Curiel TJ.. Considerations for successful cancer immunotherapy in aged hosts. Exp Gerontol. 2018;107:27-36. 10.1016/j.exger.2017.10.002. [DOI] [PubMed] [Google Scholar]
- 20. Tomihara K, Curiel TJ, Zhang B.. Optimization of immunotherapy in elderly cancer patients. Crit Rev Oncog. 2013;18(6):573-583. 10.1615/critrevoncog.2013010591. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Hopkins AM, Rowland A, Kichenadasse G, et al. Predicting response and toxicity to immune checkpoint inhibitors using routinely available blood and clinical markers. Br J Cancer. 2017;117(7):913-920. 10.1038/bjc.2017.274. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Champiat S, Lambotte O, Barreau E, et al. Management of immune checkpoint blockade dysimmune toxicities: a collaborative position paper. Ann Oncol. 2016;27(4):559-574. 10.1093/annonc/mdv623. [DOI] [PubMed] [Google Scholar]
- 23. Mohile SG, Dale W, Somerfield MR, et al. Practical assessment and management of vulnerabilities in older patients receiving chemotherapy: ASCO guideline for geriatric oncology. J Clin Oncol. 2018;36(22):2326-2347. 10.1200/JCO.2018.78.8687. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Katz S, Akpom CA.. A measure of primary sociobiological functions. Int J Health Serv. 1976;6(3):493-508. 10.2190/UURL-2RYU-WRYD-EY3K. [DOI] [PubMed] [Google Scholar]
- 25. Caillet P, Laurent M, Bastuji-Garin S, et al. Optimal management of elderly cancer patients: usefulness of the Comprehensive Geriatric Assessment. Clin Interv Aging. 2014;9:1645-1660. 10.2147/CIA.S57849. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Extermann M, Boler I, Reich RR, et al. Predicting the risk of chemotherapy toxicity in older patients: the Chemotherapy Risk Assessment Scale for High-Age Patients (CRASH) score. Cancer. 2012;118(13):3377-3386. 10.1002/cncr.26646. [DOI] [PubMed] [Google Scholar]
- 27. Presley CJ, Dotan E, Soto-Perez-de-Celis E, et al. Gaps in nutritional research among older adults with cancer. J Geriatr Oncol. 2016;7(4):281-292. 10.1016/j.jgo.2016.04.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Hurria A, Mohile S, Gajra A, et al. Validation of a prediction tool for chemotherapy toxicity in older adults with cancer. J Clin Oncol. 2016;34(20):2366-2371. 10.1200/JCO.2015.65.4327. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Magnuson A, Sedrak MS, Gross CP, et al. Development and validation of a risk tool for predicting severe toxicity in older adults receiving chemotherapy for early-stage breast cancer. J Clin Oncol. 2021;39(6):608-618. 10.1200/JCO.20.02063. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Welaya K, Loh KP, Messing S, et al. Geriatric assessment and treatment outcomes in older adults with cancer receiving immune checkpoint inhibitors. J Geriatr Oncol. 2020;11(3):523-528. 10.1016/j.jgo.2019.05.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Harris PA, Taylor R, Thielke R, et al. Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42(2):377-381. 10.1016/j.jbi.2008.08.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Baker PS, Bodner EV, Allman RM.. Measuring life-space mobility in community-dwelling older adults. J Am Geriatr Soc. 2003;51(11):1610-1614. 10.1046/j.1532-5415.2003.51512.x. [DOI] [PubMed] [Google Scholar]
- 33. Nashed A, Zhang S, Chiang CW, et al. Comparative assessment of manual chart review and ICD claims data in evaluating immunotherapy-related adverse events. Cancer Immunol Immunother. 2021;70(10):2761-2769. 10.1007/s00262-021-02880-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Naqash AR, Ricciuti B, Owen DH, et al. Outcomes associated with immune-related adverse events in metastatic non-small cell lung cancer treated with nivolumab: a pooled exploratory analysis from a global cohort. Cancer Immunol Immunother. 2020;69(7):1177-1187. 10.1007/s00262-020-02536-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Oken MM, Creech RH, Tormey DC, et al. Toxicity and response criteria of the Eastern Cooperative Oncology Group. Am J Clin Oncol. 1982;5(6):649-655. [PubMed] [Google Scholar]
- 36. Johns AC, Wei L, Grogan M, et al. Checkpoint inhibitor immunotherapy toxicity and overall survival among older adults with advanced cancer. J Geriatr Oncol. 2021;12(5):813-819. 10.1016/j.jgo.2021.02.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Badger TA, Braden CJ, Mishel MH.. Depression burden, self-help interventions, and side effect experience in women receiving treatment for breast cancer. Oncol Nurs Forum. 2001;28(3):567-574. [PubMed] [Google Scholar]
- 38. Presley CJ, Han L, Leo-Summers L, et al. Functional trajectories before and after a new cancer diagnosis among community-dwelling older adults. J Geriatr Oncol. 2019;10(1):60-67. 10.1016/j.jgo.2018.05.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Andersen BL, McElroy JP, Carbone DP, et al. Psychological symptom trajectories and non-small cell lung cancer survival: a joint model analysis. Psychosom Med. 2022;84(2):215-223. 10.1097/psy.0000000000001027. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Presley CJ, Arrato NA, Janse S, et al. Functional disability among older versus younger adults with advanced non–small-cell lung cancer. JCO Oncology Practice. 2021;17(6):e848-e858. 10.1200/OP.20.01004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Costa PT Jr, McCrae RR.. Neuroticism, somatic complaints, and disease: is the bark worse than the bite? J Pers. 1987;55(2):299-316. 10.1111/j.1467-6494.1987.tb00438.x. [DOI] [PubMed] [Google Scholar]
- 42. Howren MB, Suls J.. The symptom perception hypothesis revised: depression and anxiety play different roles in concurrent and retrospective physical symptom reporting. J Pers Soc Psychol. Jan 2011;100(1):182-195. 10.1037/a0021715. [DOI] [PubMed] [Google Scholar]
- 43. Reiche EM, Nunes SO, Morimoto HK.. Stress, depression, the immune system, and cancer. Lancet Oncol. 2004;5(10):617-625. 10.1016/S1470-2045(04)01597-9. [DOI] [PubMed] [Google Scholar]
- 44. Kiecolt-Glaser JK, Derry HM, Fagundes CP.. Inflammation: depression fans the flames and feasts on the heat. Am J Psychiatry. 2015;172(11):1075-1091. 10.1176/appi.ajp.2015.15020152. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45. Mohile SG, Mohamed MR, Xu H, et al. Evaluation of geriatric assessment and management on the toxic effects of cancer treatment (GAP70+): a cluster-randomised study. Lancet. 2021;398(10314):1894-1904. 10.1016/S0140-6736(21)01789-X. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46. Li D, Sun CL, Kim H, et al. Geriatric Assessment-Driven Intervention (GAIN) on chemotherapy-related toxic effects in older adults with cancer: a randomized clinical trial. JAMA Oncol. 2021;7(11):e214158. 10.1001/jamaoncol.2021.4158. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data underlying this article will be shared on reasonable request to the corresponding author.