A study of the efficacy and toxicity outcomes of extended durvalumab dosing in patients with stage III unresectable non-small cell lung cancer (NSCLC) during the COVID-19 pandemic

Abstract

Background

Durvalumab following chemoradiation in unresectable stage III non-small cell lung cancer (NSCLC) has led to improved outcomes. The schedule of administration has been determined by pharmacokinetic studies. This study evaluates real-world efficacy and safety outcomes of extended dosing (ED) vs. standard dosing (SD) of durvalumab.

Methods

Stage III NSCLC patients treated at the Cancer center of Southeastern Ontario with consolidative durvalumab from March 2017-December 2020 were included. Patient characteristics and outcomes were evaluated through retrospective review. Comparisons were made using chi-square and t-tests. Kaplan-Meier curves were used to analyze overall survival (OS).

Results

A total of 35 patients were included; 15 (43%) switched to ED. Distant recurrence rates were higher in the ED group (53% vs. 20%, p = 0.07), with no differences in the sites of disease recurrence. A similar proportion of patients were alive in the ED vs. SD group (93% vs. 80%, p = 0.3), with no significant difference in OS. There were less grade 3 or greater immune-related adverse events in the ED group (0% vs. 20%). Treatment discontinuation occurred in 47% vs. 50% in the ED vs. SD groups, respectively, owing to toxicity in 20% of patients in the ED group vs. 40% in the SD group.

Conclusions

Extended dosing has similar efficacy and toxicity to standard dosing; however, there was a higher rate of toxicity necessitating discontinuation in the SD group, which may have impacted the clinical decision-making to switch to ED. Our data is limited by a small sample size and should be further validated in larger cohorts.

Introduction

The standard of care for unresectable stage III non-small cell lung cancer (NSCLC) patients who responded to concurrent chemoradiation has been established by the PACIFIC trial, which demonstrated an overall survival benefit for patients who received the PD-L1 inhibitor durvalumab as consolidation therapy [1], [2], [3]. The trial used a weight-based regimen of 10 mg/kg delivered intravenously (IV) every two weeks. Weight-based durvalumab regimens have been compared to fixed-dosing regimens including 750 mg IV every 2 weeks (standard dosing; SD) and 1500 mg IV every 4 weeks (extended dosing; ED), and a similar median steady-state exposure was observed [4]. However, in Canada, the regimen largely used in clinical practice continues to be the weight-based regimen of 10 mg/kg every 2 weeks.

The Coronavirus disease-2019 (COVID-19) pandemic has led to more than 6 million deaths worldwide [5]. Cancer patients are amongst the most vulnerable populations to the disease with significant morbidity and mortality rates [6, 7]. Moreover, lung cancer patients exhibited higher morbidity and mortality rates amongst all cancer patients, with mortality rates ranging from 22% to 41% [8]. Due to these risks, it became imperative to reduce the number of treatment-related visits to try and limit cancer patients’ exposure to the virus. In November 2020, the ED regimen of durvalumab was granted FDA approval for treatment of stage III NSCLC patients. This approval was in part based on data from the use of fixed-dosing of durvalumab at 1500 mg every four weeks in the CASPIAN trial in extensive stage small cell lung cancer [9]. ESMO published a proposal for the management of lung cancer during the pandemic in April 2020 which recommended the use of the ED regimen over the SD regimen of durvalumab as a “high priority” or “Tier 1″ recommendation as informed by the Ontario Health Cancer Care Ontario framework of resource prioritization and by the ESMO Magnitude of Clinical Benefit Scale (MCBS) [10]. Subsequently, the European Union and the United Kingdom approved the ED regimen in January 2021. In Canada, Cancer Care Ontario approved the ED regimens allowing their incorporation into clinical practice. However, there remains a paucity of real-world data on clinical outcomes with alternate dosing regimens, and no phase III randomized controlled trials have evaluated ED durvalumab regimens against SD regimens. The aim of this study is to assess the real-world efficacy and toxicity of extended vs. standard dosing of consolidative durvalumab following chemoradiation in patients with unresectable stage III NSCLC.

Methods

Patient and treatment characteristics

This retrospective analysis included patients diagnosed with histologically confirmed stage III NSCLC treated with consolidative durvalumab at the Cancer center of Southeastern Ontario (CCSEO; Kingston, Ontario, Canada) between March 2017 and December 2020. SD was defined as durvalumab 10 mg/kg IV every two weeks, whereas ED was defined as durvalumab 20 mg/kg IV every four weeks. Institutional ethics approval was obtained before study commencement. Patient characteristics, treatment details, and outcomes were evaluated through chart review. Patient and pathologic characteristics included age, gender, Eastern Cooperation Oncology Group (ECOG) performance status (PS), smoking status, TNM stage at diagnosis (AJCC 8th edition), histology, EGFR/ALK/KRAS mutation status (using Oncomine Comprehensive Assay version 3 Next Generation Sequencing) and PD-L1 Tumor Proportion Score (TPS) evaluated using the 22C3 IHC pharmDx assay. The date at which the first dose of durvalumab was administered was recorded, as well as the number of cycles received. For patients who switched to the ED regimen, the date of the switch, and the number of cycles received pre- and post-switch were recorded. The date of distant recurrence, location of distant disease recurrence, and number of cycles to death/recurrence were collected. Immune-related adverse events (irAEs) included any toxicity identified by the physician as immune-related whether clinically or radiographically. Specifically, pneumonitis included both immune-mediated and radiation-induced. Toxicity data included the date of the irAE, grade of irAEs according to the Common Terminology Criteria for Adverse Events (CTCAE) version 5.0, whether treatment was withheld or continued, and whether treatment was ultimately restarted. If treatment was permanently discontinued, the reason for treatment discontinuation, including toxicity and progression, was recorded.

Study outcomes

The primary outcomes were overall survival (OS) and rate of irAEs. Secondary endpoints included rate of distant disease recurrence and site of distant disease recurrence. OS was defined as the time from initial diagnosis to death from any cause. The rate and sites of distant disease recurrence were compared between the ED and SD groups. Rates of irAEs were recorded based on type and grade and compared between both groups.

Statistical methods

All statistical analyses were conducted using IBM SPSS Statistics version 27 (Armonk, NY, 2021). Descriptive analyses included frequencies and percentages for categorical data, including baseline patient, tumor, and treatment characteristics, and the number, rate, and grade of irAEs. Comparisons of the ED and SD groups were completed using the Pearson's chi-square test, or Fisher's exact test in the event of cells with fewer than 5 cases. OS was assessed via Kaplan-Meier plots and differences were analyzed using the log-rank test. The designated threshold of statistical significance was 0.05, and no adjustment was made for multiple comparisons.

Results

Patient and treatment characteristics

A total of thirty-five patients were included in the final analysis. Fifteen patients (43%) switched to the ED schedule and the remaining twenty patients (57%) continued on the SD schedule. The unadjusted median follow-up time was 17.4 months (25th and 75th quartiles 12.8, 28.5). The median follow-up time could not be adjusted for mortality since only 5 of 35 patients died, and it was therefore undefined by the Kaplan-Meier process. Baseline characteristics for both groups are summarized in Table 1 . The only statistically significant difference in baseline characteristics was age, with a higher proportion of younger (<65 years old) patients in the ED group vs. the SD group (60% vs. 25%, p = 0.036). A majority of patients were female (73% in the ED group vs. 55% in the SD group), had non-squamous histology (80% in the ED group vs. 60% in the SD group), and had stage IIIA/B disease (73% in the ED group vs. 85% in the SD group, respectively). Twenty-nine patients (83%) had a reported PD-L1 status: 4 had PD-L1 TPS <1% (11%), 9 had a PD-L1 TPS of 1–49% (26%) and 16 had a PD-L1 TPS of ≥50% (46%). Sixteen patients (46%) had a KRAS mutation, and none had an EGFR or ALK mutation.

Table 1.

Baseline characteristics.

	SD group (N = 20) N (%)	ED group (N = 15) N (%)	Overall population (N = 35) N (%)	P-value
Age  ≥65  <65	15 (75) 5 (25)	6 (40) 9 (60)	21 (60) 14 (40)	0.04*
Gender  Male  Female	9 (45) 11 (55)	4 (27) 11 (73)	13 (37) 22 (63)	0.27
ECOG  0  1  ≥2	11 (55) 9 (45) 0 (0)	6 (40) 9 (60) 0 (0)	17 (49) 18 (51) 0 (0)	0.38
Stage  III  IIIA  IIIB  IIIC  Other/unknown	2 (10) 8 (40) 9 (45) 0 (0) 1 (5)	1 (6.7) 5 (33.3) 6 (40) 3 (20) 0 (0)	3 (8.6) 13 (37) 15 (43) 3 (8.6) 1 (2.8)	0.30
Histology  Squamous  Non-squamous	8 (40) 12 (60)	3 (20) 12 (80)	11 (31) 24 (69)	0.21
Mutations  EGFR  ALK  KRAS   G12C   Non-G12C  Non-mutated/unknown	0 (0) 0 (0) 8 (40) 5 (25) 3 (15) 12 (60)	0 (0) 0 (0) 8 (53) 5 (33) 3 (20) 7 (47)	0 (0) 0 (0) 16 (46) 10 (29) 6 (17) 19 (54)	0.51
PD-L1 status  ≥50%  1–49%  <1%  Unknown	7 (35) 5 (25) 2 (10) 6 (30)	9 (60) 4 (27) 2 (13) 0 (0)	16 (46) 9 (26) 4 (11) 6 (17)	0.14

The median number of cycles received in the ED group before and after the switch were 9 (0–19) and 5 (1–11) cycles, respectively. The median number of cycles delivered in the SD group was 12 (1–26). The treatment course of all patients on study is summarized in Fig. 1 .

Fig. 1.

Overview of treatment course for all study patients.

Efficacy

Distant recurrence rates were higher in the ED group compared to the SD group (53% vs. 20%, p = 0.07). The most common site of distant recurrence in the ED group was brain (n = 4, 27%) followed by liver (n = 2, 13%) and bone (n = 2, 13%). In the SD group, the most common site of recurrence was brain (n = 2, 10%) followed by liver and adrenal gland (n = 1, 5% each). There were no statistically significant differences in the sites of distant disease recurrence between groups (Fig. 2 ). At the time of data cut-off, 14/15 patients were alive in the ED group vs. 16/20 in the SD group (93% vs. 80%, p = 0.37). Overall survival was not significantly different between groups (p = 0.29; Fig. 3 ). Age did differ between the ED and SD groups, so a second Kaplan-Meier analysis was undertaken to assess age category and OS. This was very close to statistical significance (p = 0.050), so a third Kaplan-Meier analysis was undertaken to assess a four-level grouping consisting of age <65 years for ED and SD, and age ≥65 years for ED and SD. With both factors taken into account simultaneously, there were no significant differences overall across the four groups (p = 0.263, data not shown).

Fig. 2.

Sites of distant disease recurrence by group.

Fig. 3.

: Overall survival for patients treated with consolidative durvalumab in the standard dosing group (blue) vs. extended dosing group (green), p = 0.29.

Toxicity

The most common irAEs in the overall population included pneumonitis (43%), dermatitis (8.6%) and thyroid dysfunction (8.6%). The full list of irAEs is provided in Table 2 . Pneumonitis (any grade) occurred in 33% of patients in the ED group vs. 50% of patients in the SD group, whereas grade 3 or higher pneumonitis accounted for 0% vs. 15%, respectively. Rare yet serious irAEs such as myocarditis and nephritis occurred in three patients in the SD group and none were reported in the ED group. There were no grade 3 irAEs in the ED group, whereas 4/20 patients (20%) experienced grade 3 irAEs in the SD group. There were no grade 4 irAEs and no treatment-related deaths in either group. Treatment discontinuation occurred in 7/15 patients (47%) in the ED group vs. 10/20 patients (50%) in the SD group. Treatment was discontinued due to immunotoxicity in 3/15 patients (20%) in the ED group vs. 8/20 patients (40%) in the SD group, and treatment was discontinued due to progression in 4/15 patients (27%) in the ED group vs. 2/20 (10%) patients in the SD group. The median time to toxicity in the SD group was 1.5 months vs. 2.8 months in the ED group (p = 0.06). All patients in the ED group were re-challenged with durvalumab at a later date compared to only 33% of patients in the SD group (p = 0.08). These findings are summarized in Table 3 .

Table 2.

Initial immune related adverse events (irAE) by system.

	SD group		ED group before switch		ED group after switch
Type of irAEs – N (%)	Grade 1–2	Grade ≥3	Grade 1–2	Grade ≥3	Grade 1–2	Grade ≥3
Pneumonitis*	7 (35)	3 (15)	3 (20)	0 (0)	2 (13)	0 (0)
Colitis	1 (5)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)
Thyroid Dysfunction	3 (15)	0 (0)	1 (7)	0 (0)	0 (0)	0 (0)
Fatigue	0 (0)	0 (0)	1 (7)	0 (0)	0 (0)	0 (0)
Dermatitis	1 (5)	0 (0)	2 (13)	0 (0)	0 (0)	0 (0)
Nephritis	1 (5)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)
Myocarditis	0 (0)	2 (10)	0 (0)	0 (0)	0 (0)	0 (0)
Xerostomia	0 (0)	0 (0)	0 (0)	0 (0)	1 (7)	0 (0)

^⁎Includes both immune-related and radiation-induced pneumonitis.

Table 3.

Total immune related adverse events (irAE) by grade, treatment discontinuation and rechallenge rates.

	SD group (N = 20)	ED group (N = 15)	P-value
Rate of irAEs by grade – N (%)  Grade 1–2  Grade 3 or higher	13 (65) 5 (25)	10 (67) 0 (0)	0.13
Rate of treatment discontinuation - N (%)  Overall  Due to irAE  Due to progression	10 (50) 8 (40) 2 (10)	7 (47) 3 4	0.16
Median time to first irAE in months	1.5	2.77	0.06
Re-challenge rate	4 (33)	4 (100)	0.08

Discussion

Extended dosing of durvalumab was incorporated into clinical practice for the management of unresectable stage III NSCLC patients during the COVID-19 pandemic to help reduce the number of treatment-related visits and potential exposure to COVID-19. However, data supporting the use of alternate dosing regimens of durvalumab is largely driven by pharmacokinetic studies, with a paucity of randomized controlled trials comparing extended vs. standard dosing schedules [11]. Although treatment adverse effects and the financial impact of treatment are frequently studied outcomes in oncology, the time patients spend in coordination and receipt of care is an understudied outcome and can have a great impact on patients [12]. The so called “time toxicity”, defined by Gupta et al. as the amount of time pursuing cancer-directed therapy, can be significant [12]. Extended dosing strategies provide a convenient and attractive alternative for cancer patients who are spending a significant amount of time in cancer centres receiving treatment and can also have a potential impact on reducing health care costs associated with treatment delivery. This makes the comparison of efficacy and toxicity outcomes associated with ED vs. SD regimens an important question to be addressed. In this retrospective single-center study, we evaluated real-world efficacy and toxicity outcomes in stage III unresectable NSCLC patients treated with either ED vs. SD regimens of consolidative durvalumab.

There were no significant differences in efficacy outcomes between the ED and SD groups in our study. Despite a trend towards higher distant recurrence rate in the ED group (53% vs. 20%, p = 0.07), there was no statistically significant difference in OS between both groups at the time of data cut-off. OS results in this study are congruent with other studies in this space, which have reported similar OS between extended dosing and standard dosing groups. The Hijmering-Kapelle et al. study reported that for NSCLC patients treated with consolidative durvalumab with a median follow up of 54.6 weeks, median OS and median progression free survival (PFS) were not reached in both the SD and ED groups (p = 0.98 and p = 0.94, respectively) [13]. A study by Denault et al. evaluated 4-weekly vs. 2-weekly dosing of consolidative durvalumab in stage III NSCLC patients and reported that median OS was not reached in both groups, but 12-month survival rates were similar (85.2% vs. 88.4% in the ED and SD groups, respectively; p = 0.55) ¹⁴.

In the durvalumab arm of the PACIFIC trial, 20.4% of patients had any new distant disease recurrence [11]. Our data was similar with approximately 20% of patients who had distant disease recurrence in the SD group. Distant disease recurrence was higher in the ED group, at 53% in our study; however, this was not statistically significant (p = 0.07). In other retrospective analyses looking at stage III NSCLC patients receiving consolidation durvalumab, there have been no significant differences in distant disease recurrence rates noted between ED and SD groups [13, 14]. For instance, in the Denault et al. study, distant disease recurrence occurred in 17% vs. 31.6% in the ED vs. SD groups [14].

To our knowledge, our study is the first to include recurrence data by disease site for alternate dosing regimens. The most common site of distant recurrence was the brain in both the ED and SD groups (27% vs. 10%, p = 0.37). We observed no statistically significant differences in the sites of distant disease recurrence between both groups; however, our data was limited by small patient numbers. This highlights the need for further studies exploring the effect of dosing schedules on disease-specific sites of recurrence.

In our study, the incidence of irAEs (of any grade) in the SD group was 75%, as compared to 96.8% in the durvalumab arm of the PACIFIC trial. Four out of 15 patients (20%) in the SD group experienced grade 3 irAEs, with no grade 4 irAEs or treatment-related deaths. Overall, this is similar to the rate of grade 3 or 4 events reported in the PACIFIC trial (29.9%) [1]. Pneumonitis (including radiation pneumonitis) of any grade occurred in 33.9% of patients in the durvalumab arm of PACIFIC (with 3% of these patients having grade 3 or 4 pneumonitis) [1]. Pneumonitis occurred in 43% of the overall population in our study, with 50% in the SD group and 33% in the ED group. Grade 3 pneumonitis occurred only in the SD group, at 15%. This may explain why a greater proportion of patients in the SD group (40%) discontinued treatment due to toxicity, as compared to the PACIFIC trial (15.4%) [1].

The incidence of irAEs reported in the ED group was 67% (n = 10); all of which were grade 1 or 2. In the Mac et al. study, there was a non-statistically significant trend towards a lower rate of irAEs in the ED vs. SD group (12.5% vs. 31.3%, p = 0.39). Similar trends have been reported in other studies, with a trend towards less all-grade, and grade 3 or higher, toxicities in ED groups. In the subset of patients with stage III NSCLC treated with durvalumab in the Hijmering-Kapelle et al. study, there were 79 all-grade irAEs in the ED group (n = 49) vs. 26 irAEs in the SD group (n = 15), with grade 3 or higher irAEs accounting for 7.5% of patients in the ED group vs. 11.5% of patients in the SD group (p = 0.58) ^13. In the Denault et al. study, all-grade irAEs occurred in 55.8% vs. 58.7% of stage III NSCLC patients treated with durvalumab in the ED vs. SD groups, and grade 3 or higher irAEs occurred in 11.5% vs. 12.7%, respectively [14].

Treatment discontinuation due to irAEs occurred in a higher proportion of patients in the SD group of our study (40% vs. 20%), and only 33% of those patients were re-challenged with durvalumab. This is in contrast to the ED group where all patients were re-challenged with durvalumab. These findings are consistent with the Hijmering-Kapelle et al. study which reported a higher rate of permanent discontinuation of immunotherapy in the SD group. In the overall population treated with durvalumab, nivolumab or pembrolizumab, permanent treatment discontinuation due to toxicity occurred in 12.5% in the SD group vs. 4.3% in the ED group, whereas short-term interruptions in treatment due to toxicity occurred in 15.4% vs. 13.6% in the ED vs. SD group respectively [13]. In the durvalumab alone subset, treatment was discontinued due to toxicity in 5% in the SD group vs. 3% in the ED group [13]. Taken together, our data are consistent with the findings of other small retrospective studies, suggesting at least a similar toxicity profile of the ED regimen in comparison to the SD regimen [13], [14], [15].

Limitations

Our study has limitations. Given the retrospective nature of the analysis, important factors such as socioeconomic or psychosocial factors, among others, were not included as this data was not consistently documented in the patient chart. Notably, the current study is limited by the small number of patients in the SD and ED cohorts, as well as a short follow up interval. Further studies with larger cohorts of patients and a longer follow up interval are needed to validate our findings. Also, there was a higher rate of toxicity necessitating discontinuation of durvalumab in the SD group. This might have impacted the clinical decision-making to switch to ED, as patients switched to the ED schedule were likely to have an ongoing response and favorable side effect profile, thereby contributing to a possible selection bias. Moreover, less frequent visits associated with the ED schedule may have contributed to toxicities going undetected for longer, thereby affecting the time to toxicity.

Conclusions

This study suggests that extended dosing of durvalumab is associated with similar efficacy to standard dosing. Despite a non-statistically significant trend towards higher distant recurrence rates in the ED group, there were no statistically significant differences in the sites of distant disease recurrence or OS. Furthermore, extended dosing did not appear to be associated with an increase in toxicities compared to standard dosing.

These findings contribute to a small body of evidence investigating the important clinical question of alternative dosing schedules of immune checkpoint inhibitors. Given the small sample size and short follow up, these findings warrant further validation in larger cohorts.

While this study has its limitations, the data taken in aggregate with several other small studies, suggests that there are no major safety signals for adopting extended dosing regimens into clinical practice, especially at a time where the benefit from limiting a patient's exposure to COVID-19 cannot be undermined. We believe that the impact the COVID-19 pandemic has had on cancer care delivery will be long lasting; therefore, there is an ongoing interest in further evaluating the role of alternate dosing regimens, not only of immune checkpoint inhibitors but also for other therapeutic agents.

Funding Statement

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Consent statement

Patient consent was obtained for use of patient information for research purposes. Institutional ethics approval was obtained before study commencement.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.