FormalPara Key Points

There was a lack of effective and durable treatments for patients with HER2+ metastatic gastric cancer who had received a prior trastuzumab-based regimen before 2020.

Patients incurred substantial financial burden after progression on or after first-line trastuzumab.

Novel therapies such as trastuzumab deruxtecan may help improve patient outcomes and mitigate disease burden.

1 Introduction

Gastric cancer accounts for 1.4% of all new cancer cases in the United States (USA), with an estimated 26,560 new cases and 11,180 deaths in 2021 [1]. Despite a longstanding decline in the global incidence of gastric cancer, the absolute number of new cases per year is increasing as the population ages [2]. In the USA, about 60% of gastric cancer cases are diagnosed at advanced stages, associated with poor prognosis with a 5-year survival rate not exceeding 20% [3, 4]. Overexpression or amplification of human epidermal growth factor receptor 2 (HER2+) occurs in approximately 20% of gastric cancer and is therefore an important therapeutic target [5, 6].

Information on real-world treatment outcomes is scarce for patients with HER2+ metastatic gastric cancer (mGC) following progression on first-line (1L) trastuzumab in combination with chemotherapy. Although addition of trastuzumab to chemotherapy in the 1L setting has shown a survival benefit over chemotherapy alone [7], trastuzumab continuation beyond progression in advanced gastric cancer has failed to show clinical benefit in patients with HER2+ mGC in clinical trials [8, 9]. Other HER2-targeted agents, including pertuzumab, lapatinib, and ado-trastuzumab emtansine (T-DM1), have historically failed to show benefits in patients with gastric cancer [10,11,12]. Barzi et al. conducted a retrospective analysis using electronic medical records (EMR) data from 2013 to 2018 and found marked heterogeneity in options for second-line (2L) treatment in mGC [13]. HER2-targeted treatment for patients with HER2+ disease following 1L trastuzumab-based regimens was not available for mGC until the approval of fam-trastuzumab deruxtecan-nxki (T-DXd) in 2021. Before then, ramucirumab monotherapy or in combination with chemotherapy, and pembrolizumab monotherapy were indicated for the treatment of mGC patients with disease progression, on or following prior lines of chemotherapies [14].

Gastric cancer imposes significant clinical and economic burden on the healthcare system. Abraham et al. estimated a mean all-cause healthcare cost of US$16,242 per patient per month (PPPM) among patients with advanced or metastatic gastric cancer [15], and it was estimated the total national cost for hospitalizations with gastric cancer increased from US$1.79 billion in 2003 to US$1.96 billion in 2014 [16]. However, economic data specific to HER2+ patients, especially those who received treatment following 1L trastuzumab-based regimens, are limited. The purpose of this study was to characterize the unmet medical need post-1L trastuzumab, including real-world treatment patterns, healthcare resource use (HRU), and costs in patients with HER2+ mGC prior to the approval of T-DXd in 2021. Such information can provide benchmarks for healthcare resource use and costs of prior therapies, which may be used to estimate the value of emerging treatments.

2 Methods

2.1 Data Source

This retrospective analysis was conducted using data from the IQVIA PharMetrics® Plus Database between October 2014 and September 2019. The database is comprised of adjudicated medical and pharmacy claims for more than 190 million unique commercially insured enrollees across the USA and is considered representative of the national commercially insured population aged 65 years and under. All data are de-identified and compliant with the Health Insurance and Portability Act. Approval from an ethics committee was not required.

2.2 Study Sample

Patients were required to have a gastric cancer diagnosis (International Classification of Disease 9th edition, Clinical Modification [ICD-9-CM] codes 151.0-151.9 or ICD-10-CM codes C16.0-C16.9) followed by a secondary malignancy diagnosis (ICD-9-CM codes 196.xx-198.xx or ICD-10-CM codes C77.xx-C79). The date of the first secondary malignancy diagnosis was defined as the mGC diagnosis date. Patients with a diagnosis of any primary cancers other than gastric cancer during 6 months prior to their mGC diagnosis were excluded from the study sample.

Patients must have received trastuzumab in 1L on or after the mGC diagnosis. First line began on the date of the first infusion or fill of a systemic antineoplastic agent after the mGC diagnosis date, including trastuzumab and other cancer agents filled or infused within the first 30 days. Trastuzumab in combination with capecitabine following trastuzumab in combination with platinum-based chemotherapy was considered a maintenance therapy and not identified as a new line of therapy (LOT), unless the trastuzumab + capecitabine-based regimens contained new drugs that were not included in the prior regimen. The index date was defined as the date of discontinuing 1L trastuzumab-based regimen (Fig. 1). Discontinuation was defined as a treatment gap ≥ 60 days for all treatments in a LOT or the initiation of a new treatment [17]. Patients were required to have continuous eligibility starting from the 6 months prior to mGC diagnosis and were followed until the end of data availability or continuous eligibility. The study period was from the index date to the earliest of the end of data availability or end of continuous eligibility.

Fig. 1
figure 1

Study design schema. 1L first line, GC gastric cancer, mGC metastatic gastric cancer

Full size image

2.3 Study Outcomes and Statistical Analysis

2.3.1 Patient Characteristics

Patient demographic and clinical characteristics, including age, sex, geographic region, time from mGC diagnosis to discontinuation of 1L trastuzumab-based regimens, National Cancer Institute Comorbidity Index (NCICI) score [18], and metastatic sites (bones, lungs, brain, liver, and peritoneum/retroperitoneum) were summarized during the baseline period. Mean and standard deviation or median were used to summarize continuous variables, while percentages were used to summarize categorical variables.

2.3.2 Treatment Patterns After First-Line (1L) Trastuzumab-Based Regimens

Treatment patterns of 2L were reported by proportion of patients receiving each regimen after 1L trastuzumab-based regimens. Patients were categorized into mutually exclusive groups: (1) trastuzumab-based regimen; (2) paclitaxel + ramucirumab; (3) fluorouracil + irinotecan; (4) pembrolizumab; and (5) other regimens. Time to treatment discontinuation of 1L trastuzumab-based regimens, time from 2L initiation to discontinuation, and time from 2L initiation to initiation of 3L treatment was evaluated using Kaplan-Meier analyses.

2.3.3 Healthcare Resource Use (HRU) and Costs

The numbers of all-cause outpatient visits (including outpatient infusion visits and other outpatient visits not involving infusion), inpatient admissions, emergency department (ED) visits, and other visits (skilled nursing facility, hospice, durable medical equipment, and dental or vision care) were analyzed from the date of the discontinuation of 1L trastuzumab-based regimens to the end of the study period and were reported on a PPPM basis for two groups: patients who did and did not receive 2L treatment. Additionally, HRU during 2L treatment was reported for patients who had evidence of subsequent treatment, stratified by regimen groups.

All-cause healthcare costs, expressed on a PPPM basis, were reported after the discontinuation of 1L trastuzumab-based regimens and while patients were on 2L treatment, respectively. The cost categories (in 2020 US dollars) comprised inpatient, ED, outpatient (including outpatient infusion visits and other outpatient services), and pharmacy.

3 Results

3.1 Patient Characteristics

Of 298 patients with HER2+ mGC who received 1L trastuzumab-based regimens, 190 who discontinued 1L treatment were included in this study (Onine Supplementary Material (OSM) Fig. 1). The mean (standard deviation (SD)) age of the study cohort was 56.83 (9.50) years and 74% were male (Table 1). The median follow-up from the index date was 7.11 months. The mean (SD) NCICI score at baseline was 1.83 (1.65). The most common metastatic site was liver (47.89%), followed by lungs (23.16%) and bones (20.53%).

Table 1 Patient characteristics
Full size table

Kaplan-Meier analysis of the 298 patients showed a median (95% confidence interval (CI)) time to discontinuation of 1L trastuzumab-based regimens of 5.09 (4.40, 5.82) months, similar to the eight cycles of trastuzumab administration reported in the TOGA trial [7]. After 12 months of initiation of 1L trastuzumab-based regimens, only 17.08% of patients remained on treatment (OSM Fig. 2).

3.2 Treatment Patterns

Among 190 patients who discontinued 1L trastuzumab-based regimens, 136 (72%) initiated subsequent treatment while 54 (28%) did not. The median follow-up from 2L initiation was 6.34 months. The most common 2L therapy was trastuzumab-based regimens (n = 69; 50.74%), followed by ramucirumab + paclitaxel (n = 27; 19.85%), fluorouracil + irinotecan (n = 8; 5.88%), and pembrolizumab (n = 8; 5.88%) (Fig. 2). The 2L trastuzumab-based regimens consisted of trastuzumab in combination with chemotherapy (n = 57; 82.61%), trastuzumab in combination of other targeted therapy (n = 7; 10.14%), and trastuzumab monotherapy (n = 5; 7.25%). The median (95% CI) time to 2L discontinuation was 2.37 (2.07, 2.83) months, and less than 13% of patients remained on 2L after 12 months of 2L initiation (Fig. 3a). The median (95% CI) time from 2L initiation to 3L initiation was 6.28 (5.09, not reached) months, with 60% of patients initiating 3L after 12 months of 2L initiation (Fig. 3b).

Fig. 2
figure 2

Distribution of 2L regimens post-1L trastuzumab-based regimens (N = 136). a Other included regimens such as carboplatin + paclitaxel, ramucirumab, fluoropyrimidine + oxaliplatin ± leucovorin, irinotecan, and lapatinib + capecitabine

Full size image
Fig. 3
figure 3

Time from initiation of 2L treatment to treatment discontinuation and initiation of 3L treatment. 2L second line, 3L third line

Full size image

3.3 HRU and Costs

After discontinuing 1L treatment, patients who did not receive subsequent treatment were less likely to have outpatient infusion visits (14.81% vs. 98.53%) and ED visits (31.48% vs. 44.12%), compared to those who received subsequent treatment (Table 2). Patients without subsequent treatment had lower mean number of outpatient visits PPPM (3.84 vs. 5.71), but higher mean number of inpatient admissions (0.31 vs. 0.20), ED visits (0.19 vs. 0.13) and longer mean length of inpatient stay (5.02 vs. 2.84 days).

Table 2 HRU before and after discontinuation of 1L trastuzumab-based regimens
Full size table

During 2L treatment (mean = 3.11 months), HRU varied by subsequent treatment (Table 3). The mean (SD) number of outpatient visits, inpatient admissions, and length of inpatient stay (in days) were highest among patients receiving fluorouracil + irinotecan; 9.69 (9.54), 0.17 (0.36), and 6.96 (5.71), respectively, although the sample size is small (n = 8). Patients receiving trastuzumab-based regimens had the highest mean (SD) number of ED visits per month (0.13 (0.37)), and patients receiving paclitaxel + ramucirumab had the highest mean (SD) number of other visits (0.27 (1.31)).

Table 3 HRU on 2L treatment
Full size table

Mean (SD) healthcare cost during baseline period before discontinuation of 1L trastuzumab-based regimens for the entire cohort of 190 patients was $22,735 ($13,290) PPPM. Baseline costs were similar between patients with subsequent treatment and those without (Fig. 4; OSM Table 1). After discontinuing 1L treatment, patients who received subsequent treatment incurred higher mean (SD) healthcare costs PPPM than those without subsequent treatment ($25,178 ($16,398) vs. $14,812 ($16,370)), primarily driven by the higher cost of outpatient infusions ($12,134 ($11,822) vs. $1020 ($3498)). The mean (SD) inpatient cost PPPM among patients who did not receive subsequent treatment increased from $4893 ($8165) to $9297 ($12,842) after discontinuation of 1L treatment.

Fig. 4
figure 4

Mean healthcare costs PPPM during baseline and follow-up periods. ED emergency department, IP inpatient, OP outpatient, PPPM per patient per month, USD United States dollars. The bolded numbers represent total costs. Other OP costs include OP costs not for treatment infusion. Other costs included costs for skilled nursing facility, hospice, durable medical equipment, and dental or vision care

Full size image

Mean (SD) PPPM cost on 2L treatment was $30,838 ($23,863) among all patients receiving 2L treatment, the majority of which was attributable to outpatient infusion costs ($22,262 ($20,898)) (Fig. 5; OSM Table 2). Similar to the trend for HRU, the costs of 2L varied by regimen. On average, the mean (SD) PPPM costs were $17,016 ($12,869) for patients receiving fluorouracil + irinotecan, $27,273 ($15,121) for trastuzumab-based regimen, $37,964 ($32,215) for paclitaxel + ramucirumab, and $49,934 ($38,945) for pembrolizumab. Outpatient infusion costs were the primary driver of total healthcare costs across all regimens.

Fig. 5
figure 5

Mean healthcare costs PPPM on 2L treatment (total and by regimen). ED emergency department, IP inpatient, OP outpatient, PPPM per patient per month, USD United States dollars. The bolded numbers represent total costs. Other OP costs include OP costs not for treatment infusion. Other costs included costs for skilled nursing facility, hospice, durable medical equipment, and dental or vision care

Full size image

4 Discussion

This retrospective analysis of US claims data characterized treatment patterns and healthcare costs following 1L trastuzumab-based regimens in patients with HER2+ mGC. Consistent with other retrospective studies in gastric cancer [13, 19,20,21], considerable variability was observed in 2L treatments in the studied cohort, suggesting lack of standard treatment strategies following 1L treatment during the study period. The study revealed that trastuzumab in combination with chemotherapy, targeted therapy, or by itself, was used in more than half of the patients in later lines of therapy. During the study period, no HER2-targeted treatments were approved or recommended for 2L or subsequent therapy. In the absence of alternative targeted therapy, oncologists might choose to reuse trastuzumab in combination with different chemotherapy. However, the clinical efficacy of trastuzumab continuation beyond 1L has not been demonstrated in any studies [8]. At the time of this study, FOLFIRI (leucovorin + fluorouracil + irinotecan), FOLFOX (leucovorin + fluorouracil + oxaliplatin), pembrolizumab monotherapy, and ramucirumab with or without paclitaxel were recommended for 2L and beyond. These regimens were used less frequently than trastuzumab-based treatment in our findings. The median treatment duration for 2L treatment was 2.37 months, comparable to median PFS observed from two other Phase II trials evaluating the continuous use of trastuzumab beyond progression in HER2+ advanced gastric cancer: 2.4 months with irinotecan plus trastuzumab [8] and 3.7 months with paclitaxel plus trastuzumab [9]. The modest duration of 2L treatment or median PFS suggests a lack of durable efficacy of treatments available before 2020 for patients with HER2+ mGC who had received a prior trastuzumab-based regimen. Approximately 28% patients did not receive subsequent systemic treatment after 1L trastuzumab-based regimens. While the exact reason for the absence of subsequent treatment could not be ascertained in the absence of clinical information, it is possible that these patients were unfit for additional treatment, or declined cancer treatment due to uncertain clinical benefits and expected toxicity likely to reduce quality of life.

Patients with HER2+ mGC incurred substantial financial burden to the healthcare system after progression on or after 1L trastuzumab treatment. The mean total monthly cost following discontinuation of 1L trastuzumab-based regimens was $22,093, ranging from $14,812 for patients without subsequent treatment to $25,178 for patients receiving subsequent treatment. Inpatient care was the primary driver of healthcare costs in patients without subsequent therapy, while infusion cost was the primary driver for those with subsequent treatment. The PPPM costs while patients were on 2L treatment varied by regimen; patients receiving pembrolizumab or ramucirumab in combination with paclitaxel incurred much higher costs than the others. During the study period, pembrolizumab was only approved for the treatment of patients with recurrent locally advanced or metastatic gastric cancer whose tumors express PD-L1 and who had disease progression on or after two or more prior systemic therapies [22]. In the absence of biomarker results from claims data, we could not ascertain PD-L1 status of patients receiving pembrolizumab. However, the low proportion of pembrolizumab used observed in this study was consistent with the low PD-L1 testing rate (14.8%) in advanced gastric cancer [23]. In general, outpatient services, including infusion and non-infusion services, drove the high 2L treatment cost across regimens.

The lack of durable treatments and the high economic burden observed in this study highlight the need for new effective targeted treatments for HER2+ mGC patients who progress on or after trastuzumab-based regimens. T-DXd, a HER2-directed antibody and topoisomerase inhibitor conjugate, was approved on January 15, 2021 by the FDA for this patient population based on findings from the multicenter, open-label, randomized DESTINY-Gastric01 trial [24]. Treatment with T-DXd was associated with a significantly higher objective response rate (51% vs. 14%), longer progression-free survival (PFS) (median: 5.6 vs. 3.5 months) and longer OS (median: 12.5 vs. 8.4 months; all p < 0.05) compared with chemotherapy [24]. Because the trial was conducted in South Korea and Japan, its generalizability to the Western patients was evaluated in the phase II, single-arm, open-label DESTINY-Gastric02 trial, where T-DXd demonstrated consistent efficacy with a confirmed objective respond rate of 38.0%, median progression-free survival of 5.6 months, and median OS of 12.1 months [25]. The newly approved 2L anti-HER2 agent T-DXd may shift the treatment paradigm and help mitigate the unmet needs of patients with HER2+ mGC in the USA who progress after 1L trastuzumab-based regimens. Future studies are warranted to confirm the effectiveness of T-DXd in the real-world setting.

Several limitations should be considered when interpreting the current study findings. First, because HER2 test results are not available in claims database, anti-HER2 regimens were used as a proxy for HER2+ status. Second, in the absence of complete clinical information, an algorithm was developed to define LOTs. It was not possible to determine if a treatment discontinuation was a temporary interruption or permanent discontinuation. The data did not capture reasons to explain why some of the patients did not receive subsequent treatment. Finally, because the IQVIA PharMetrics® Plus Database primarily captures claims from commercial insurance, generalizability of our study findings needs to be confirmed with patients with other types of insurance, such as Medicare Fee-For-Service, or uninsured patients.

5 Conclusions

Patients with HER2+ mGC who had received a prior trastuzumab-based regimen had unmet clinical needs, evidenced by heterogeneity in practice patterns and a lack of effective and durable treatments prior to 2020. The healthcare system faced substantial burden, primarily resulting from high drug acquisition costs. Novel therapies such as T-DXd may help improve patient outcomes and mitigate the disease burden. The present findings complement existing clinical trial data to provide insight into the real-world outcomes related to treatment options for HER2+ mGC and a baseline estimate to inform economic evaluations and health technology assessments. Future real-world studies assessing the economic implications for the novel anti-HER2 therapies or immunotherapies would be useful to better inform healthcare providers and administrators on value-based decisions.