Discussion
Conventional cytotoxic chemotherapy has played an important role in the comprehensive treatment of cancer. However, it has limitations associated with side effects and requires long break periods, generally ranging from 2 to 3 weeks between treatment cycles, during which time tumors sometimes grow.18 MC refers to the chronic administration of chemotherapeutic agents at relatively low, minimally toxic doses, without prolonged drug-free breaks.1
Here, we report the results of a single-center retrospective analysis of children with relapsed or refractory solid tumors who received a metronomic regimen based on low-dose continuous oral administration of anticancer drugs. In line with most published studies on pediatric MC, we have found benefits in terms of disease stabilization, defining “responders” as tumors that presented CR, PR, and SD.8,10–12,14 The overall response rate was 61%, and 11 patients were defined as responders. Even among the patients who started MC in PD status, 45% presented with PR or SD.
Seven out of 11 responders turned to PD during MC, and 5 of them switched back to conventional chemotherapy. Of 5 responders who were in PD status at the start of MC, all died of the disease after tumor re-progression. However, the responders had prolonged survival times and maintained good QOL when compared to non-responders, with a median duration of response of 8 months (range, 3.3–102.3 months).
To date, most previous clinical trials and retrospective analyses of pediatric MC have focused on relapsed or refractory progressive tumors.1,11 Although only a few cases achieved CR with MC, most reports indicate that some clinical benefit, including PR or SD, occurs in 25–67% of cases.
Robinson et al. performed a phase 2 trial on a metronomic 5-drug regimen including celecoxib, thalidomide, fenofibrate, and alternating 21-day cycles of etoposide and cyclophosphamide.10 Of the 96 patients who were evaluated for tumor response, 1 achieved CR, 12 achieved PR, 36 achieved SD, and 46 showed PD; the response rate was 51% according to the same definition as in our study.
Zepletalova et al. reported the results of a prospective study for the COMBAT metronomic regimen that included low-dose daily temozolomide, etoposide, celecoxib, vitamin D, fenofibrate, and retinoic acid.8 Of 62 patients with measurable disease, 40% improved and achieved remission or SD. The 2-year overall survival (OS) in all patients was 43.1%, which is statistically prolonged compared to historical controls (median OS was 15.4 months vs. 3.0 months). Our study suggested that MC can contribute to tumor dormancy and prolonged survival time, as expected from these previous larger-scale studies.
On the other hand, some clinical studies have reported discouraging data. For example, Pramanik et al., who performed a randomized control study (RCT) for altering oral etoposide and cyclophosphamide with celecoxib and thalidomide, reported that 6-month PFS was not improved when compared to placebo.12
However, subgroup analyses revealed some encouraging results; patients who had received the drugs for at least 9 weeks without tumor progression had prolonged PFS. In other words, the initial treatment response after the start of MC is important for improving the prognosis. This is consistent with our analysis indicating that most of the responders showed radiographic response at the first imaging performed 1–3 months after the initiation of MC.
Because of the variety of patient backgrounds and previous treatments, it is difficult to discuss the tolerance of MC. However, our report has shown that MC is at least associated with fewer infectious episodes and transfusion dependence. In previous metronomic trials,4–10 dosage regulation was permitted to reduce clinically significant toxicities or to avoid chemotherapy discontinuation. Most of our patients also received adjustment of oral dosage, and although hematological toxicity was almost inevitable, the episodes of FN in individual patients were significantly fewer when compared to conventional chemotherapy (Fig. 4: P < 0.01, paired t-test). This is considered an important advantage of MC because FN usually requires intravenous antibiotic treatment and hospitalization.
In Japan, cytotoxic chemotherapy is usually performed in the setting of continuous hospitalization in case of infection or the need for a blood transfusion. Consequently, pediatric patients spend less time with their families and hence sometimes experience loneliness. Moreover, if they are of school age, they have to give up their regular school life.19,20 Treatment options on an outpatient basis are important for patients with a poor long-term prognosis.
In our retrospective analysis, AEs directly related to QOL, such as appetite for food, sense of fatigue, or performance status, were not evaluated sufficiently. However, MC can induce tumor dormancy without forcing hospitalization. Rather than simply seeking to prolong life, MC has the potential to provide a continuous treatment environment for the patient that prioritizes their life in the family and society.
Another expected role of MC is maintenance therapy.13In our study, two patients received MC in CR, and both patients completed the therapy without relapse and maintained CR. It is not clear whether the success of treatment in these two cases depended on MC. In the past, studies of MC limited to solid tumors in remission were rare. Senerchia et al. performed a controlled study on operated osteosarcomas comparing a regimen of methotrexate, anthracycline, cisplatin (MAP regimen), with MAP combined with MC, but no advantage was demonstrated in the combination therapy.21 However, as mentioned later, patients with osteosarcomas are not suitable for verifying the advantage of MC.
Bisogno et al., in an RCT on children with RMS, reported that the addition of metronomic maintenance therapy with daily oral cyclophosphamide and weekly intravenous vinorelbine resulted in a significant increase in overall and event-free survival.17 Although this trial was targeted to primary non-metastatic disease, efficacy in a palliative setting has also been reported.22
Focusing on adult oncology, MC as a maintenance therapy has been tried in various combinations, including for colorectal cancer, breast cancer, and ovarian cancer.13 Although most of these studies are small, single-arm observations, a good safety profile and promising efficacy have been demonstrated. Of course, delayed complications (such as secondary malignancies due to long-term maintenance use of alkylators) should be considered,23 and the appropriate treatment duration should also be discussed. Metronomic maintenance therapy for pediatric tumors requires further studies and accumulation of evidence, including treatment duration and delayed complications.
Recently, many molecular-targeted drugs have been developed that are becoming an important part of cancer treatment, either in combination therapy with anticancer drugs or as maintenance therapy. However, the identified actional mutations do not cover all cancers. Pediatric tumors, in particular, are known to have fewer target gene mutations than adult tumors.24 MC may be a useful option as maintenance therapy for patients with cancers that are inappropriate for targeted agents.
This study has some limitations, as it was a small-scale, single-center retrospective analysis. The survival time after MC initiation was not directly compared with that of conventional chemotherapy or no treatment. Moreover, tumor pathogenesis, primary site, clinical course, disease status before MC, combination therapy, and details of the metronomic regimen were variable. Although univariate analysis could not reveal significant confounding factors (expect for tumor response), there may have been a bias in the disease groups of the patients.
Previous studies have indicated that low-grade glioma, NBL, and RMS are likelier to respond to MC than high-grade glioma and osteosarcoma,12,21,22,25 which could affect our response rate. It may be difficult to eliminate all of these limitations or to conduct comparative studies with a specific disease population, considering the rarity of pediatric cancer.
The literature on pediatric MC mostly includes case reports, retrospective analyses, and phase I and II clinical studies performed on heterogeneous populations. Our study, although even smaller with a more limited and variable patient population, demonstrated that MC can play an active role in palliative and maintenance therapy for relapsed or refractory pediatric solid tumors. These results are in line with the results of previous reports, and it would be worthwhile to accumulate experience and clinical data on MC for pediatric patients.