What is the real value of metronomic chemotherapy in the treatment of gastrointestinal cancer?

1. Introduction

Conventional chemotherapy (CT) is administered at maximum tolerated doses (MTDs) required to attain tumoristatic/tumoricidal effects. Unfortunately, it requires lengthy drug-free breaks due to myelosuppression, which allows time for tumor-bed recolonization by chemo-resistant clones, leading to treatment failure. Metronomic chemotherapy (MTC), on the other hand, is the more frequent administration of chemotherapeutic drugs at less toxic dosages (below MTD) over specific intervals, minimally requiring treatment interruption [1]. Easy access, remarkable tolerability and cost-effectiveness as well as notable activity in resistant tumors are all features that rendered MTC an intriguing research topic in oncology, including gastrointestinal cancers (GI) [2]. MTC is tested in subsequent therapy settings, as maintenance therapy, or in upfront settings in frail CT-ineligible patients. It has a wide spectrum of anticancer effects including anti-angiogenesis, cancer stem-cell reduction, and cellular immune-modulation [2] (Figure 1). Its direct and indirect effects on tumor cells, immune cells and tumor microenvironment via calibration of circulating regulators such as cytokines, growth factors and angiogenic proteins may explain MTC activity in later lines of resistant tumors [2].

Figure 1. Mechanism of action of metronomic chemotherapy in malignancies

In this editorial, we discuss relevant clinical data of MTC used in variable subtypes of gastrointestinal cancers.

2. MTC by site of primary disease

2.1. Colorectal cancer (CRC)

Capecitabine (Cap), tegafur-uracil (UFUR), doxifluridine (5′-DFUR), and S-1 (tegafur/gimeracil/oteracil) are all oral forms of fluoropyrimidines [3]. Fluoropyrimidines are considered to be the backbone of chemotherapeutic regimens in CRC [3]. Metronomic dosing of these drugs, most commonly Cap (mCap), was evaluated in several trials.

In the phase-III Nordic ACT2 trial, KRAS-mutated mCRC patients with no progression >18 weeks of 1^st-line treatment were randomized to receive bevacizumab (BVZ) every 3 weeks or mCap (500 mg bid) in metronomic dosing [4]. The outcomes were relatively similar. 3-month progression-free survival (PFS) rates reached 75% and 66.7% (p = 0.58), and median overall survival (mOS) reached 26.4 and 28 months (p = 0.13) in the BVZ and mCap groups respectively [4]. This established non-inferiority of mCap as maintenance therapy.

In CAIRO-3, 558 mCRC patients without disease progression >6 cycles of first-line CAPOX-BVZ were randomized to receive mCap (625 mg/m2 bid) continuously plus BVZ 7.5 mg/Kg every 3 weeks versus observation [5]. The experimental arm showed higher median PFS1 (mPFS1) and PFS2 (mPFS2) (8.5 vs 4.1 months, p < 0.0001 & 11.7 vs 8.5 months, p < 0.0001 respectively) [5]. A trend toward improved mOS (25.9 months vs 22.4, p = 0.06) was noted without quality of life compromise [5], indicating mCap effectiveness and tolerability in combination with other agents. On the other hand, in MOMA trial (phase-II), the intensification of maintenance MTC with the addition of cyclophosphamide to mCap and BVZ following induction FOLFOXIRI-BVZ showed no advantage [6].

In another retrospective study, UFUR (400 mg bid) maintenance following adjuvant FOLOFOX for stage-III CRC improved 5-year OS rates over observation (86.8% vs 68.5%, p = 0.01 respectively). Patients receiving >5 months of treatment showed the greatest benefit with 5-year OS rate reaching 90.3% [7]. This once more proved MTC efficacy as maintenance therapy.

Metronomic S-1 (mS-1) (80 mg/m2/d on d3-7,10–14,17-21) in combination with weekly intravenous irinotecan (CPT11) (60 mg/m2) was studied as first-line treatment of 45 mCRC patients [8]. The response rate (RR) was 48.9%, mPFS 8.1 months, and mOS 20.9 months. The regimen was well tolerated with mild toxicities [8]. In another study, 5′-DFUR (800 mg/m2/d, 5 days a week) and CPT11 (40 mg/m2 weekly for 3 of 4 weeks) resulted in RR of 36%, mPFS of 187 days, and mOS of 452 days in 45 mCRC patients [9]. In a third study, metronomic CPT11 continuously infused daily for 3 weeks out of 4 (escalated doses in 3 groups: 1.4 mg m⁻² day⁻¹/DI = 9.8 mg m⁻² week⁻¹; 2.8 mg m⁻² day⁻¹/DI = 19.6 mg m⁻² week⁻¹; 4.2 mg m⁻² day⁻¹/DI = 29.4 mg m⁻² week⁻¹) seemed to be well-tolerated and achieved disease control in 20% of mCRC patients refractory to chemotherapy including irinotecan [10]. This indicated potential clinical activity of MTC in resistant tumors despite significantly low therapy-dosing.

2.2. Pancreatic cancer (PC)

In one study, induction S-1 plus metronomic-dose cisplatin (4 mg/m2 on d1-5,8–12,15-19,22–26/28d cycles) followed by S1 maintenance resulted in a RR of 17%, and a disease control-rate (DCR) of 90% (27/30), with no grade 4 toxicities [11]. It proved to be a very tolerable, less toxic regimen in comparison to others in literature [11].

The combination of low-dose weekly nab-paclitaxel (60 mg/m²) and oxaliplatin (50 mg/m²), plus continuous infusion 5FU (180 mg/m2/d, d1–14/28–35d cycle with BVZ (5 mg/m² d1,15) was retrospectively studied in 65 patients with advanced PC [12]. The regimen proved to be effective and safe with a RR of 49% and a DCR of 81%. mOS was impressive at 19 months (82% living >1 year) [12].

Furthermore, mCap (1500 mg/d) studied in 22 pretreated patients with pancreato-biliary cancer resulted in 33% SD, with higher survival rates in fit patients [13].

All these studies present MTC as a possible tolerable therapeutic option in PC.

2.3. Gastro-esophageal cancer (GEC)

A retrospective series of 47 patients, 25 with progressive GEC receiving mCap (1500 mg/d continuously), proved treatment safety with PR achieved in 1 patient and SD in 6 patients [13]. In another prospective study, mCap (1000 mg/d, d1–28/35) was evaluated in 45 elderly pretreated GEC patients [14]. The RR was 20.9%, the median time to progression (mTTP) was 3.6 months and mOS was 7.6 months [14]. The toxicity profile was acceptable with <10% grade 3 toxicities and no grade 4 toxicities [14].

Another study with metronomic oral etoposide (50 mg/m²) and intravenous 5FU (200 mg/m²) d1-14/28-d cycles in 28 GEC patients resulted in an overall RR of 50%, with an mTTP of 4.5 months and mOS of 9.5 months [15].

All these studies show efficacy and safety of MTC in GEC.

2.4. Hepatocellular carcinoma (HCC)

In one phase-II trial, two cohorts of 59 CT-naïve and 31 pretreated HCC patients received mCap at a dose of 500 mg/m2 bid [16]. PFS reached 6 and 3.3 months, whereas OS was of 14.5 and 9.8 months respectively (16). In another retrospective study, patients failing sorafenib received best supportive care (BSC) (n = 112) or mCap (500 mg bid) (n = 117). In the mCap arm, both median post-sorafenib survival (PSS) and OS from sorafenib initiation were better than BSC (9.9 vs 5.8 months, p = 0.001; and 17.5 vs 12 months, p = 0.034, respectively) [17]. These results were endorsed in a multicentric study reporting better mOS for patients on mCap (500 mg bid) vs BSC after sorafenib discontinuation: 12 vs 9 months; HR 0.45 (p = 0.013) [18].

Thereby, mCap can be a safe and efficacious therapeutic option in HCC patients failing 1^st-line sorafenib [18].

In an Egyptian randomized-controlled trial of 40 HCC patients, mCap showed a favorable toxicity profile over doxorubicin in the first-line setting, without a significant difference in TTP (3.4 vs 3.1 months) nor OS (10.2 vs 9.6 months) [19], indicating the possibility of replacing doxorubicin with mCap. Nevertheless, mCap-BVZ combination in the 1^st-line setting had limited antitumoral activity (RR 9%, DCR 52%), and efficacy (PFS 2.7 months, OS 5.9 months) [20], decreasing the enthusiasm for its use in 1^st-line setting.

In one study, the use of metronomic UFUR in combination with sorafenib in 53 HCC patients showed some antitumoral activity (RR 8%, DCR 57%), with a mPFS of 3.7 months, a mOS of 7.4 months, and a similar toxicity profile versus sorafenib alone [21]. Thalidomide-metronomic UFUR combination was also tested, resulting in a RR of 9%, a DCR of 33%, mPFS of 1.9 months, and mOS of 4.6 months [22].

Therefore, in HCC, MTC is well tolerated and can be effective alone or in combination; however, its activity remains rather limited.

2.5. Bile duct cancer

In a phase-I trial, 6 patients failing 2^nd-line CT received weekly low-dose paclitaxel (40 or 50 mg/m2). DCR was 83.3% and mOS from the third-line treatment was 9 months [23], providing a possible treatment choice in this poor-prognosis category of patients.

2.6. Neuroendocrine neoplasms (NEN)

In a study combining metronomic continuous-infusion 5FU (200 mg/m2/day) for ≥6 months with long-acting-release (LAR) octreotide in 29 untreated well-differentiated NEN patients, radiologic RR was 24%, DCR was 93%, biochemical RR was 48% and symptomatic RR was 60%. Therapy was well tolerated [24]. Another phase-II trial studied the XELBEVOCT regimen (mCap 1000 mg x2/d plus biweekly BVZ and octreotide) in 45 patients (32 were CT-naïve). The treatment was well tolerated. Radiologic RR was 17.8% and DCR was 82%, with better response in tumors of pancreatic origin. Biochemical and symptomatic responses were observed in 52.9% and 82.3% of cases respectively [25].

These studies provide insight that even in NEN, MTC can have a role, and is well tolerated.

3. Expert opinion

MTC has proven itself a valid therapeutic option with impressive biologic rationales, good tolerability, and potential clinical benefits. However, available data regarding MTC use in GI malignancies are insufficient to undoubtedly incorporate it in treatment paradigms. Several pitfalls regarding evidence quality exist. Most published data are based on small samples lacking randomization, with the highest-quality evidence coming from few phase-III trials. Furthermore, studied populations are quite heterogeneous, undermining the possibility of generalizable clinical knowledge. Appraisal of MTC in the later lines of treatment by which cancer cells acquired resistance to several previously utilized antineoplastic agents would plunge the likelihood of success of MTC. Besides, multiple questions arise regarding dosing-schedules and variable definitions of MTC regimens. This lack of a unified MTC definition entails the necessity for more pharmacokinetic and pharmacodynamic studies properly defining the low active range of drug concentrations over specific time-intervals that do not increase toxicity nor compromise the quality of life. Some studies have attempted the use of mathematical calculations to establish optimal metronomic dosing; yet, further validation is still required [26].

For mCRC, the role of mCap ± BVZ has shown encouraging results in palliative settings and maintenance therapy for 1^st-line responders [27]. In HCC, mCap improved OS compared to BSC beyond sorafenib, with better results in good performance patients (18). Even in PC, GEC and NEN, MTC remains to be a valid therapeutic option [28]. Due to its low toxicity, MTC could be particularly helpful for the elderly and frail [29], accounting for a significant proportion of patients with GI malignancies who are MTD-CT-ineligible. Finally, MTC is a cost-effective therapeutic alternative [30], specifically in low-income countries, given its low cost, easy-access, and less need for infrastructure compared to MTD CT (2).

Efforts should focus on conducting large prospective trials (Table 1) [31] (ideally phase-III trials) incorporating MTC alone or in combination with synergetic drugs, to attain high-quality evidence supporting its use, specifically in earlier treatment settings [32]. It is important to identify patient-subpopulations for whom MTC would be best and establish the ideal predictive biomarkers that correlate with clinical outcomes [33]. Broader awareness of the potential of MTC should be acknowledged, with more investigations implemented in pursuit of the integration of MTC in cancer treatment algorithms.

Table 1. Active clinical trials with metronomic CT in GI malignancies

Tumor Site	Clinical Trial	Study Phase	Study Population	Number of Patients	Treatment	Outcomes	Study Status
Gastric Cancer	NCT02855788	Phase II	Stage III/IV GC patients who previously received Oxaliplatin and 5-FU in other MTD regimens	40	Paclitaxel 60 mg/m2, Oxaliplatin 50 mg/m2, Leucovorin 20 mg/m2, and 5-FU 425 mg/m2 IV weekly	Primary outcome: Response Rate Secondary outcomes: AEs- PFS- OS	Recruiting
	NCT02007148	Phase II	Advanced gastric/GEJ adenocarcinoma patients who progressed on 1 prior line of chemotherapy including a fluoropyrimidine and a platinum agent	51	Capecitabine 500 mg twice a day, orally, continuously Docetaxel 60 mg/m2 i.v. over 1 h on day 1 cycles repeated every 3 weeks until disease progression	Primary outcome: Tumor response Secondary outcomes: PFS-OS-AEs-time to treatment failure	Recruiting
Colorectal Cancer	NCT04534218	Phase II	Stage IV colorectal cancer patients in progression after previous standard treatments (5FU, CPT11, oxaliplatin, anti-VEGF and anti-EGFR therapy if KRAS and NRAS WT	49	Regorafenib: 80 mg dailyx1 week,120 mg dailyx1 week,160 mg daily for the third week of the first cycle), then 80, 120 or 160 mg daily dose according to toxicity+ Capecitabine: 625 mg/m^2/orally twice daily continuously until progression; Cyclophosphamide: 50 mg dailyx6 months; Aspirin 75 mg daily until progression	Primary Outcome: ORR	Not yet recruiting
	NCT04425564	Phase II	Stage IV colorectal cancer with un-resectable metastases with no prior treatment for metastatic disease and >6 m from adjuvant therapy	70	XELOX(oxaliplatin 130 mg/m2 + Capecitabine 1000 mg/m2 BID) vs. Metronomic low-dose Capecitabine 2000 mg daily/2 doses x 8 weeks + oxaliplatin 30 mg/m2 weeklyx8 then 2 weeks rest	Primary Outcomes: RR-toxicity-peak/ trough levels of Capecitabine Secondary Outcomes: PFS-OS	Completed, not reported
	NCT02280694	Phase II	Stage IV colorectal cancer patients who progressed following all available chemotherapy treatment lines (including chemotherapy, bevacizumab±ziv-aflibercept, and an EGFR inhibitor [if WT(wild type)-KRAS]	45	Cyclophosphamide 50 mg, 1X1/day ONLY days 1–5/week; At evening only; Capecitabine 500 mg, fixed dose of 1500 mg/day (1000 mg at morning + 500 mg at evening) days 1–5/ week; Methotrexate 2.5 mg, 1x2/day ONLY on days 6–7/week; At morning AND evening (one hour before meal); Celecoxib 200 mg, 1x2/day	Primary outcome: PFS	Completed, not reported
	NCT03158610	Phase II/III	Stage IV colorectal cancer patients who responded to 16–18 months first-line chemotherapy	25	Maintenance therapy: Capecitabine at metronomic dose of 500 mg/m2 BID QD Vs. Capecitabine 1000 mg/m2 BID d1-d14 Q3w	Primary outcome:PFS1/2 Secondary outcomes: OS-treatment related AE-QoL	Enrolling by invitation
	NCT02887365	Phase IV	Stage II colon cancer patients with MSS/MSI-low disease after complete resection of primary with no prior chemo- or radiation therapy	300	Metronomic UFT vs observation for 1 year as maintenance therapy	Primary outcome: 3-year disease free survival Secondary outcome: 5-year OS	Recruiting
Pancreatic Cancer	NCT03959150	Phase II/III	Stage II/III pancreatic cancer patients who have undergone tumor resection, and received at least 6 cycles of Gemcitabine + Capecitabine adjuvant therapy	231	Maintenance metronomic Capecitabine at 500 mg/m2 QD versus observation for 1 year	Primary outcome: 1-year disease free survival Secondary outcomes: OS-RFS-AEs-exploratory biomarkers	Recruiting
	NCT03387098	Phase Ib/II	Advanced pancreatic adenocarcinoma patients who progressed on or after SoC chemotherapy	173	Multi-modal treatment including MCT + NANT immunotherapy (Aldoxorubicin HCl, ALT-803, ETBX-011, GI-4000, haNK, Avelumab, Bevacizumab, Capecitabine, cyclophosphamide, fluorouracil, Leucovorin, nab-paclitaxel, omega-3-acid ethyl esters, oxaliplatin, SBRT)	Primary outcomes: AEs in phase Ib-ORR in phase II Secondary outcomes: ORR in phase Ib-PFS-OS-duration of response-DCR	Active, not recruiting
Gastro-entero-pancreatic Neuroendocrine Tumors	NCT02736500	Phase I/II	Inoperable or metastatic gastro-entero-pancreatic NEN with a positive 18FDG PET/CT	37	28 GBq 177Lu-DOTATATE 5 cycles of 5.5 GBq (150 mCi) each, up to the total cumulative activity of 28 GBq (750 mCi) 177Lu-DOTATATE Or 22 GBq 177Lu-DOTATATE 6 cycles of 3.7 GBq (100 mCi) each, up to the total cumulative activity of 22 GBq (600 mCi) 177Lu-DOTATATE (depending on risk stratification)	Primary outcomes: ORR-toxicity Secondary outcomes: PFS	Active, not recruiting
	NCT02838342	Phase II	Un-resectable NETs with Ki67 ≤ 10%	28	Metronomic cyclophosphamide (50 mg/d x 6 months; Interferon-alpha 3 months of treatment from the 4th cycle metronomic cyclophosphamide: subcutaneous injection, initial dose of 3 MIU 3x/wk	Primary outcome: rate of patients with immune response	Active, not recruiting

Abbreviations: GC: gastric cancer; 5FU: 5fluoro-uracil; MTD: maximum tolerated dose; IV: intra-venous; AE: adverse event; PFS: progression-free survival; OS: overall survival; GEJ: gastro-esophageal junction; ORR: overall response rate; RR: response rate; WT: wild-type, QoL: quality of life; BID: twice daily; Q3w: every 3 weeks; MSS/MSI: micro-satellite stable/micro-satellite instable; QD: daily; RFS: recurrence-free survival; MCT: metronomic chemotherapy; SoC: standard of care; DCR: disease control rate; SBRT: stereotactic body radiation therapy; NEN: neuro-endocrine neoplasms; NET: neuro-endocrine tumors

Declaration of interest

O Abdel-Rahman has served on an advisory board for Eisai, Eli Lilly and Company and Roche. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Reviewer Disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.