Hemostatic palliative radiotherapy for gastric cancer: A literature review

Highlights • Palliative radiotherapy is rarely used for the treatment of gastric cancer.• Standard treatment for inoperable advanced gastric cancer is not established yet.• Low doses of approximately BED10 of 30 Gy (EQD2 of 24 Gy) exert hemostatic effects.• Three prospective trials have been reported and more prospective trials are needed.


Introduction
Gastric cancer is commonly diagnosed at advanced stages and often accompanied by bleeding.Its management includes endoscopic therapy, surgery, chemotherapy, and radiotherapy.However, a consensus on treatment strategies has not been established.Conversely, radiation therapy is noninvasive and its utility has been reported in previous retrospective studies.
Palliative radiotherapy (RT) is rarely used for the treatment of gastric cancer; however, pre-or postoperative RTs are commonly administered in Europe and the United States [1][2][3][4][5], as it is noninvasive and a highly effective radical treatment.
Generally, palliative RT dose is lower than the radical RT dose, with lower chances of side effects.Thus, the purpose of RT should be properly understood.Bleeding in stomach cancer cannot be completely stopped using palliative RT.The efficacy of hemostatic RT against bleeding from the head and neck, bronchi, rectum, bladder, and uterus has already been reported [6].
The standard treatment for inoperable advanced gastric cancer has not yet been established because the treatment method is selected after assessing the irradiation method and results of retrospective studies.Moreover, only a few prospective studies are currently available.This review describes the current evidence available on palliative hemostasis irradiation for gastric cancer and the direction of future approaches for palliative irradiation in cases of gastric cancer [7][8][9][10][11][12][13][14][15][16][17][18].

Material and methods
A tree of selections and criteria is shown in Fig. 1.
Literature search was carried out on the PubMed database, selecting data between 2007 to present using the following relevant keywords: "Gastric Cancer," "Bleeding," "Radiotherapy," and "Palliation" (Fig. 1).Surveillance performed within Japan was excluded from this review [17].Articles written in languages other than English and case reports were also excluded.Previously, review articles have been reported, however, no prospective studies are included (prospective studies have been reported since 2019

Setting the end-point for gastric cancer hemostasis
As with all types of palliative medicine, determining the end-point of the treatment is essential, including symptom relief, increased time spent at home owing to discharge, and survival duration [8][9][10][11][12].
Because several previous retrospective studies were available for review, medical record changes based on the quality of life (QoL) and survival were evaluated.However, the results often do not reflect the patients' real prognosis because several limitations were associated with these retrospective studies, making it difficult to set an end-point for diseases with a short prognosis.We therefore describe the evaluation outcomes separately in each report.

Results
The response shows the rate at which the hemostatic effect was obtained (Table 1).Only prospective studies by Saito et al. estimated QoL changes, which were not considered in this literature review.Evaluation of strictures was not covered in this review because of the inconsistent assessment criteria; however, several reports have demonstrated successful stenosis release.

Dose calculations
The linear-quadratic model is commonly used in radiotherapy units and allows for easy analysis of equivalent dose for different fractionations.Equivalent dose in 2 Gy fractions (EQD2) is the dose obtained using a 2 Gy fraction dose, which is biologically equivalent to a total dose D (dose) given with a fraction dose of Gy.Moreover, the EQD2 values may be added in separate parts of the treatment plan and may be adapted in fraction doses other than 2 Gy.In our 20 reviews, biologically effective dose 10 (BED10) was the most common at 30-40 Gy, and EQD2 ( 10) was mostly at 30 Gy.
Lee et al. reported that a high response rate was achieved when BED regimens (α/β ratio = 10) had exceeded 36 Gy [8], whereas some published reports like the study by Tey demonstrate good therapeutic   [19].The primary outcomes were symptom response rates.Secondary outcomes covered overall survival (OS).Fifty patients were accrued in their study, and median survival duration was 85 days.Their primary outcomes were the alleviation of fatigue, nausea, and pain subscales of the EORTC QLQ-C30.Those results were seen in 50 %, 28 %, and 44 % of patients at the end of RT and in 63 %, 31 %, and 50 % of them 1 month after the RT.Their study shows that palliative gastric RT is effective and well-tolerated, while it allowed alleviation of fatigue, dysphagia, and pain at the end of the RT and 1 month after the completion of the RT in a significant proportion of patients.
Saito et al. also emphasized on the QoL without fixing the prescription dose (8 Gy single dose, 20 Gy/5 fractions, and 30 Gy/10 fractions) [20].Their treatment response rates were 47 %, 53 %, and 49 % at 2, 4, and 8 weeks of follow-up, respectively.However, neither response nor BED (α/β = 10) predicted OS.Univariate cox model showed that BED was not significantly associated with rebleeding.This result is similar to those of previous retrospective reports, indicating that BED was not related to bleeding control.However, they found that the per-protocol response rate increased to 90 % during the 8-week follow-up.They concluded that the above protocol is useful, although the appropriate BED was unknown.
Conversely, in Tanaka et al.'s study, repeated irradiation was examined with an emphasis on bleeding, using a fixed prescription of 20 Gy/5 fx for the entire stomach [21].The response rate of initial the RT was 80 % (25/31), and 6 of the 25 patients underwent repeated irradiation; all 6 were responders (100 %) for the prescribed 15 Gy/5 fx for partial irradiation of the stomach.The median OS was significantly different among the entire cohort and between one-time and repeated irradiation groups (91, 76, and 112 days, respectively).
Thus, even in a prospective study, successful treatment depends on the end-point used in the trial because the interpretation of results becomes more complicated based on the QoL, survival time in palliative irradiation, and recurrence-free survival time.

Discussion
The purposes of palliative treatment widely varied in the assessed studies and are as follows: 1) QoL improvement, 2) extension of survival time, and 3) extension of the asymptomatic period.However, determining the aspects of palliative treatment that are beneficial to a patient is currently impossible [24,25].Hemostatic irradiation is extremely beneficial to patients and should be investigated in detail whenever possible.However, there were some limitations in investigating the effect of hemostatic RT, because the end-points determined in each report were different, e.g.: 1) The parameters of a successful treatment, such as hemoglobin (Hb) levels are preset (several studies use an Hb level of 8.0 g/dl), and hemostatic treatment, for endoscopic treatment and surgery, is generally provided when Hb levels are < 8.0 g/dl.Hemostatic effect is defined as treatment success if Hb remains > 8 g/dl 1 month post-irradiation.However, the definition of hemostatic effects varies depending on the paper, and most studies set the Hb value to ~8 g/ dl.RT is also indicated for hemostatic irradiation when Hb levels consistently have values < 8.0 g/dl.However, if Hb levels return to ≤ 8 g/dl post-transfusion, the bleeding did not spontaneously stop.The evaluation method covers an easy-to-understand parameter (Hb level of 8 g/dl) that assesses the extent of the Hb value increase before irradiation or whether the Hb value at pretreatment can be maintained.In retrospective examinations, end-point has not been set; thus, an accurate assessment is considered difficult.Moreover, the patient can be discharged after the bleeding has been stopped.Furthermore, gastrointestinal tract stenosis can be ameliorated using RT.
2) The survival time cannot be compared between the treated and untreated groups.3) QoL can be determined through the parameters that quantitatively scores a patient's condition, which is difficult to conduct retrospectively using QoL improvement score [22].Moreover, accurately evaluating QoL is complicated as the data of retrospective studies are inferred from medical records and scored accordingly.Because of the difficulties associated with these studies, prospective studies should be also included in the future review.As of 2022, three prospective studies have been conducted, and we present a summary of previous reports in the literature shown below.
Conversely, several positive outcomes of treatment have been reported, such as those by Song et al. [23], who evaluated the clinical outcomes of endoscopic hemostasis for bleeding in patients with unresectable gastric cancer.Successful initial hemostasis was attained in 83 % of the patients, and rebleeding occurred in 28.3 % of the patients within 30 days [7,8,14,21,22].Hemostasis is achieved using a more noninvasive method than endoscopic technique or surgical bypass surgery.
Based on the reports on different treatment methods, the evidence to date indicates that hemostatic RT is sufficiently effective.The "Gastric cancer: ESMO Clinical Practice Guideline for diagnosis, treatment and follow-up" [26] only mentions preoperative RT; however, palliative RT has been proven to be effective for the treatment of gastric cancer with bleeding.Hemostatic RT should be considered in the guidelines.Symptoms, disease-free survival, OS time, and QoL may be important considerations for patients in palliative care.Although only a few reviews encompassing prospective studies have been conducted, further evidence on palliative irradiation may serve as a guide to favorable outcomes.However, conducting a comparative study on treated for bleeding and untreated patients is questionable.Reviews of previous retrospective studies demonstrated that the hemostatic effect has been found effective, and the assignment to no-treatment may be at a patient's disadvantage.The purpose of palliative irradiation is not to treat cancer but to ameliorate symptoms.Furthermore, palliative medicine should be as minimally invasive as possible, and the treatment is expected to be completed in a short time.
Tey et al. also reported no difference between doses higher and lower than 39 Gy and concluded that high doses are non-essential.Low dose also useful for patients to be treated in a short time, such as 20 Gy/5 fx [21], because patients who receive palliative irradiation often have a short survival time.Additionally, when endoscopic hemostasis cannot be achieved, surgery or radiotherapy is an option.However, radiotherapy is often more suitable than surgery for palliative medicine with a short survival prognosis.Therefore, reducing the number of hospital visits as much as possible would be useful for patients.Because a single irradiation dose of 8 Gy can also provide the desired effects, such methods are also effective for patients who have difficulty in irradiation positioning setting.For example, if a patient resides far from the hospital, they may face difficulty in shifting to a RT department because of their condition [20].As a result of these retro-and prospective studies, the established irradiation dose was controversial.Although some studies concluded that increased doses proved effective, others did not.Currently, the correlation between the median BED10 and the hemostasis rate is unclear, and if there is no correlation, a dose with high short-term efficacy should be selected.Furthermore, considering the organ at risk (OAR) dose around the stomach, if repeated irradiation becomes necessary, it would be better to lower the OARS dose for the first irradiation dose.
If the irradiation field can be observed using cone beam computed tomography, it is acceptable to reduce the planning target volume margin as the former can be narrowed and the dose administered to the liver and kidney can be reduced.Moreover, if peristalsis can be suppressed by butyl scopolamine, the accuracy will be even higher.
GRID radiotherapy is used for palliative radiotherapy to treat symptoms associated with cancer and is a radiation therapy that is sometimes administered in conjunction with conventional/threedimensional conformal radiation therapy.This method is useful when the cancer has metastasized, and the target is large.

Conclusions
Based on the evidence reported to date, it is evident that even low doses of approximately BED10 of 30 Gy exert hemostatic effects.However, only three prospective trials have been reported.Thus, in the future, more prospective studies should be performed to establish a standard treatment for palliative irradiation for gastric cancer.RT is very important in palliative care.Therefore, it is desirable to complete the treatment in a short period of time.To achieve this, trials comparing the efficacy and side effects of different fractions, such as 1, 5, or 10 fractions, are warranted.

Fig. 1 .
Fig. 1.Details of the search strategy We found 20 non-comparative observational studies including patients who received hemostatic radiotherapy for the treatment of local bleeding.
). Viani et al. reviewed 11 studies, including a prospective by Tey et al., but not by Tanaka (2020) and Saito (2021).A Abbreviations: BED, Biologically effective dose; EQD2, Equivalent dose in 2 Gy fractions; Hb, hemoglobin; OAR, Organ at risk; OS, Overall survival; QoL, Quality of life; RT, Radiotherapy.Patient Support in Radiation Oncology 2022 review article by Kawabata only examined 10 cases published since 2017.At present, no research has reviewed all reports from 2007 to 2022.Therefore, this review of retro-and prospective observational studies aimed to prepare material for future randomized trials.