Development of systemic chemotherapy starting from comparison with best supportive care: commentary on the INTEGRATE IIa phase III study

Development of active agents other than anti-angiogenic agents

Gastric cancer is the fifth most common malignancy and the third leading cause of cancer-related deaths worldwide. Its incidence varies geographically, with the highest rates observed in Eastern Asia (1,2). The standard care of unresectabe advanced or recurrent gastric and gastro-esophageal junction cancer (AGC) is systemic chemotherapy, but prognosis of AGC patients is poor.

Development of systemic chemotherapy for AGC started as the first-line treatment in early 1990’s. Three phase III trials demonstrated that a 5-fluorouracil (5-FU) containing chemotherapy regimens showed survival benefits for AGC compared with best supportive care (3-5). Although three phase III trials failed to show superiority of chemotherapy containing 5-FU plus cisplatin over 5-FU alone in terms of overall survival (OS) as the first-line treatment for AGC (6-8), doublet chemotherapy with 5-FU plus cisplatin was recognized as the standard treatment because of a higher response rate. The SPIRITS trial was the only phase III study confirming a survival benefit by adding cisplatin (9). Oral fluoropyrimidine, such as capecitabine and tegafur/gimeracil/oteracil (S-1) (10,11), and oxaliplatin replace the infusional 5-FU and cisplatin, respectively (12,13). Thus, various regimens of doublet chemotherapy with fluoropyrimidine plus platinum have been recognized as the backbone of new drug development for AGC.

In late 1990’s, new drug such as taxanes and irinotecan were developed, and a couple of phase III trials confirm survival benefits of second-line chemotherapy with taxane or irinotecan alone compared with best supportive care (14-16). In Japan, the WJOG4007 trial showed similar efficacy of weekly paclitaxel and irinotecan as the second-line treatment (17).

At the beginning of the 21st century, the door of a molecular target agent has been opened for AGC. The ToGA trial showed a survival benefit of trastuzumab, an anti-human epidermal growth factor receptor 2 (HER2) antibody, in addition to doublet chemotherapy for HER2-positive AGC, accounting for 10–20% of all AGC (18). However, most of other clinical trials developing molecular target agents other than anti-angiogenic agents for AGC failed.

In 2010’s, immune checkpoint inhibitor (ICI) and antibody-drug conjugate (ADC) have made a breakthrough in systemic chemotherapy for AGC, in both of which clinical development was started as later-line treatment. In the Attraction-2 trial, monotherapy with nivolumab showed longer OS than placebo as the third or later-line treatment, associated with longer progression-free survival (PFS) and a higher disease control rate (Table 1) (19). Subsequently, the Checkmate-649 trial and Keynote-859 trial showed survival benefits of nivolumab and pembrolizumab in combination with doublet chemotherapy as the first-line treatment for HER2-negative AGC (20,21). For HER2-positive AGC with combined positive score (CPS) >1, pembrolizumab prolonged OS in addition to trastuzumab plus doublet chemotherapy as the first-line treatment in the Keynote-811 trial (22). As the later-line treatment, trastuzumab deruxtecan (T-Dxd) showed a significantly higher response rate associated with prolonged OS compared with chemotherapy of physician’s choice in the DESTINY-Gastric01 trial (Table 1) (23). Recently, T-Dxd showed longer survival compared with weekly paclitaxel plus ramucirumab for HER2-positive AGC confirmed immediately before the second-line treatment in the DESTINY-Gastric04 trial (24). Thus, starting from the later-line, the position of these two effective agents has shifted to the earlier line. Recently, zolbetuximab, an anti-claudin 18.2 antibody, prolonged OS in combination with doublet chemotherapy as the first-line treatment for claudin18.2-positive AGC in SPOTLIGHT and GLOW trials (25,26). Thus, personalized medicine according to these biomarkers has been established for AGC. Besides these progresses, trifluridine/tipiracil (TAS-102) prolonged OS compared with placebo as the later-line treatment in the TAGS trial, associated with longer PFS and a higher disease control rate but small difference in a response rate (Table 1) (27).

Table 1. Phase III studies as the third- or later-line chemotherapy.

Study	Agent	OS			PFS		Disease control rate (%)
		Median (months)	HR (95% CI)		Median (months)	HR (95% CI)
Attraction-2	Nivolumab	5.26	0.63 (0.51–0.78)		1.61	0.60 (0.49–0.75)	40
	Placebo	4.14			1.45		25
DESTINY-Gastric01†	T-Dxd	12.5	0.59 (0.39–0.88)		5.6	0.47 (0.31–0.71)	87
	Chemotherapy‡	8.4			3.5		63
TAGS	TAS-102	5.7	0.69 (0.56–0.85)		2.0	0.57 (0.47–0.70)	44
	Placebo	3.6			1.8		14
REGARD§	Ramucirumab	5.2	0.77 (0.60–0.98)		2.1	0.48 (0.38–0.62)	49
	Placebo	3.8			1.3		23
APATINIB	Apatinib	6.5	0.71 (0.54–0.94)		2.6	0.44 (0.33–0.60)	42
	Placebo	4.7			1.8		9
Angel	Apatinib	5.8	0.93 (0.71–1.15)		2.8	0.58 (0.47–0.71)	42
	Placebo	5.1			1.8		13
INTEGRATE IIa	Regorafenib	4.5	0.68 (0.52–0.90)		1.8	0.53 (0.40–0.70)	–
	Placebo	4.4			1.6		–

^†, for HER2-positive AGC; ^‡, physician’s choice; ^§, second-line setting. Note: comparison across the clinical trials is not scientifically appropriate. AGC, unresectabe advanced or recurrent gastric and gastro-esophageal junction cancer; CI, confidential interval; HR, hazard ratio; OS, overall survival; PFS, progression-free survival; T-Dxd, trastuzumab deruxtecan; TAS-102, trifluridine/tipiracil.

Anti-angiogenetic agents for advanced gastric cancer

In the AVAGAST trial, which was the first phase III study investigating the efficacy of anti-vascular endothelial growth factor (VEGF) antibody, bevacizumab, compared with placebo as the first-line treatment for AGC, although combination chemotherapy with bevacizumab showed a higher response rate and longer PFS, there was no significant difference in OS (28). Considering from the regional difference in hazard ratios (HRs), worse in the Asian region, for OS, it was discussed that differences in independent prognostic factors and use of subsequent therapies may explain the regional difference in an OS benefit although bevacizumab showed its direct efficacy in a higher response rate and longer PFS consistently among regions. A survival benefit is usually evaluated as a HR of OS over the control arm. Considering that OS is PFS plus post-progression survival, a difference in PFS has a smaller impact on a HR of OS if post-progression is substantially long even in a randomized phase III trial in which a test arm showed longer PFS. Therefore, it would be difficult to show an OS benefit if many patients receive subsequent chemotherapy which prolong post-progression survival in a randomized phase III trial. Subsequently, in the RAINBOW trial, which was a phase III trial comparing ramucirumab, an anti-VEGF receptor (VEGFR) antibody plus weekly paclitaxel with placebo plus paclitaxel, combination with ramucirumab showed a longer OS associated with a higher response rate and longer PFS (29). Thereafter, weekly paclitaxel plus ramucirumab has been recognized as a standard chemotherapy in the second-line setting.

As monotherapy with anti-angiogenic agent, the REGARD trial showed that ramucirumab as the second-line treatment prolonged OS compared with placebo, associated with longer PFS and a higher disease control rate but a small difference in a response rate (29). The relatively higher HR for OS in the second line setting in the REGARD trial was considered to be caused by subsequent treatment. In the Chinese phase III trial, apatinib, a small-molecule tyrosine kinase inhibitor of VEGFR-2, showed longer OS compared with placebo for AGC patients with at least two lines of prior chemotherapy, longer PFS and a higher disease control rate but a small difference in a response rate (30). Although its global study (ANGEL) failed to prolong OS compared with placebo, apatinib showed longer PFS and a higher disease control rate but a small difference in a response rate (31). It was discussed that more patients in the Angel trial received subsequent chemotherapy which might dilute survival benefits of apatinib. Furthermore, in the pooled analysis of the INTEGRATE II and INTEGRATE IIa trials (32), regorafenib which is a multi-kinase inhibitor targeting angiogenic, stromal, and oncogenic receptor tyrosine kinases, with placebo for refractory AGC, regorafenib prolonged OS. While this pooled analysis of the primary endpoint seems rather unique, these two studies showed consistent results. In the INTEGRATE IIa trial, median OS was 4.5 versus 4.0 month, associated with longer PFS. These results indicate that one treatment option using regorafenib as the later-line chemotherapy is added to treatment of AGC.

The efficacy of monotherapy with an anti-angiogenic inhibitor, either monoclonal antibody or tyrosine kinase inhibitor as later-line treatment, is very consistent showing significantly longer PFS associated with a higher disease control rate and low objective response rate (Table 1). It is considered that tumor stabilization rather than tumor shrinkage resulting in longer PFS may contribute to OS prolongation. However, looking at the shape of the PFS curves and relatively higher proportions (about 50%) of patients who experience disease progression as the best response, optimal patient selection is warranted. However, many translational researches failed to establish biomarkers for selecting patients to be treated with an anti-angiogenic agent alone.

Clinical significance of later-line chemotherapy

While many guidelines refer to the second- or later lines of chemotherapy, treatment strategy including later-line chemotherapy is important to obtain long survival in clinical practice (33-35). It is recommended that available agents not used in prior treatment can be considered for subsequent-line chemotherapy in the Japanese guideline. Nowadays, many anti-tumor agents are available in various lines of chemotherapy for AGC such as ramucirumab, nivolumab, TAS-102, regorafenib for all AGC, and T-Dxd for HER2-positive AGC whose survival benefits compared with placebo have been confirmed in phase III trials. However, patient’s condition is getting worse after failure of preceding chemotherapy, and the proportion of patients who can receive the third-line chemotherapy was reported to be only 25% even in Japan practically (36). Because it is considered that only few patients can use all these active agents through all lines of chemotherapy, the choice of an optimal agent for the third-line chemotherapy is very important. However, as mentioned above, monotherapy with these agents showed compatible efficacy except for T-Dxd. Furthermore, there have been no established biomarkers for selecting the optimal agent among these active agents. Therefore, it is very difficult to select one optimal agent in this setting. Expecting the synergic efficacy among an anti-angiogenic agent, ICI and TAS-102, it is necessary to develop more effective combination chemotherapy in the third-line setting although the fraction of optimal patients for each combination is small depending on the prior chemotherapy.

Supplementary

The article’s supplementary files as

Ethical Statement: The author is accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.