Proton beam therapy for esophageal cancer compared to existing treatments, including X-ray therapy and surgery

Abstract

Esophageal cancer is one of the most difficult cancers to treat since it is often at an advanced stage at the time of symptom presentation. For locally advanced esophageal cancer, treatment options include multidisciplinary treatment such as surgery or definitive chemoradiotherapy. Surgery has a high local control rate because it involves excision of the cancer along with the surrounding organs; however, it is still highly invasive, although advances in surgery have reduced the burden on patients. On the other hand, chemoradiotherapy may also be applicable in cases in which surgery is inoperable owing to complications or distant lymph node metastasis. However, chemoradiotherapy using X-ray irradiation can cause late toxicities, including those to the heart. Proton beam therapy is widely used to treat esophageal cancer because of its characteristics, and some comparisons between proton beam therapy and X-ray therapy or surgery have recently been reported. This review discusses the role of proton beam therapy in esophageal cancer in comparison to X-ray therapy and surgery.

Key Words: Esophageal neoplasms; Prognosis; Proton beam therapy; Chemoradiotherapy; X-ray therapy; Esophagectomy; Toxicity; Quality of life; Dose volume histogram

Core Tip: Despite medical advances, esophageal cancer has a poor prognosis. Recently, there have been many reports on proton beam therapy combined with chemotherapy. Compared to X-ray therapy, proton beam therapy can reduce toxicity and may lead to long-term survival while maintaining the quality of life. Some reports suggest that it may be possible to achieve results comparable to those of surgery, although this is premised on an environment in which appropriate post-recurrence treatment can be performed. Proton beam therapy plays a significant role in the treatment of esophageal cancer in an aging society.

INTRODUCTION

Esophageal cancer (EC) is the 11^th most commonly diagnosed cancer and the seventh leading cause of cancer-related deaths in 2022[1]. The burden of EC is increasing due to an aging society, with 307.4 thousand deaths and 7.2 million disability-adjusted life-years. This is particularly problematic in Asia, where approximately 75% of EC cases and deaths occur[2]. Squamous cell carcinoma (SCC), and adenocarcinoma are the main EC pathologies. With advances in medical science, the 5-year overall survival (OS) rates of SCC and adenocarcinoma of the esophagus increased from 3.6% to 21.1% and 5.4% to 24.2%, respectively, between 1973 and 2010[3]. However, EC persists with a poor prognosis. If left untreated, the five-year OS rate is equal to or less than 10%, even including stage I cases with good general conditions where surgery is possible[4].

Therefore, radical treatment for EC is necessary, even if the patient is elderly, has comorbidities, or is in poor general condition. The standard treatment for very early-stage cancer is endoscopic resection, and for locally advanced cancer, preoperative chemoradiotherapy (CRT) followed by surgery is the standard treatment[5,6]. Conversely, the standard treatment in Japan is triplet chemotherapy (cisplatin, 5-luorouracil, and docetaxel), followed by surgery. This approach stems from clinical trials where surgery was performed after preoperative chemotherapy due to concerns about increased toxicity from radiotherapy[7,8]. CRT is indicated in cases where surgery is not possible or denied. In particular, in an aging society, many patients cannot undergo surgery because of comorbidities, even if the patient wishes to undergo surgery[5-7].

Although radiotherapy alone may be curative, it is advisable to combine it with chemotherapy whenever possible[5-7,9]. Although CRT has shown better results compared to surgery alone[10], the recurrence rate is inevitably higher than that of preoperative therapy followed by surgery[5-7]. Even if there is a local recurrence, endoscopic treatment including photodynamic therapy is possible if the disease is not locally advanced[11,12], but surgery is required if the disease progresses. Some clinical trials have shown that the mortality rate of salvage surgery is not high but is higher than that of the initial surgery and should be avoided if possible[13,14]. Other problems are late toxicity, including cardiopulmonary toxicity, which, if encountered, can result in reduced quality of life (QOL).

Recently, particle beam therapy has been used, and it is expected that damage to the surrounding organs can be reduced owing to its unique physical characteristics. As shown in Figure 1, particle beam therapy exhibits a phenomenon known as the Bragg peak, where the radiation dose rises sharply and then rapidly decreases near the end of the beam. In contrast, X-ray therapy demonstrates a gradual increase in dose after irradiation, peaking before gradually declining, which inevitably allows low-to-medium doses to spread to the surrounding tissues[15]. Proton beam therapy (PBT) and carbon-ion radiotherapy are used clinically. PBT has been widely used in combination with chemotherapy for EC, and many reports have been published on its clinical advantages and superior dose distribution compared to X-ray therapy[16-40]. Recently, a comparative study was conducted between definitive PBT and adjuvant chemotherapy followed by surgery[41]. There have also been reports of curative treatment with carbon-ion radiotherapy for EC; however, these are limited to stage I[42], and there have been no reports of curative carbon-ion radiotherapy combined with chemotherapy.

Figure 1 Schematic diagram of X-ray irradiation and proton beam therapy.

In this narrative review of treatment modalities, we explain PBT for EC, which has been increasingly reported, particularly as a curative treatment, and compare it with the existing X-ray therapy and surgery.

COMPARING DOSE-VOLUME PARAMETERS BETWEEN PROTON BEAM AND X-RAY THERAPIES

This section focuses on simulation studies that compare dose distributions, followed by a detailed comparison of actual clinical toxicity in the next section. The heart and lungs are the main organs for which dose reduction by PBT is expected, and many reports have been published on the effectiveness of this reduction[16-28]. Makishima et al[17] compared the dose volume between PBT and X-ray three-dimensional conformal radiotherapy (3DCRT). In this study, the organ volume irradiated at least xGy (VxGy), mean dose, and changes in normal tissue complication probability (NTCP), which is the probability of toxicity occurring calculated using a formula, were examined. Lung V5-20 Gy, mean lung dose (MLD), and heart V30-50 Gy were significantly lower in the PBT plan. Moreover, the NTCP was lower in the PBT plan than in the X-ray plan in all cases. In the case of the heart, the NTCP, which had a maximum of over 80%, was reduced to approximately 20% in the PBT plan; in many other cases, the probability was reduced to just a few percentages. The maximum NTCP level in the lungs was approximately 12%; however, it was reduced to only a few percentages in the PBT plan. The irradiated area depends on EC progression; however, this shows that it is possible to deliver a dose to the tumor while significantly reducing cardiopulmonary doses. Of note, among patients who received X-ray therapy, the incidence rates of grade 2 or higher radiation pneumonitis and pericardial effusion were 21% and 52.6%, respectively, compared with 0% and 4%, respectively, in patients who received PBT. In addition to 3DCRT, Hirano et al[18] compared the dose-volume parameters of PBT with those of intensity-modulated radiotherapy (IMRT), an X-ray therapy method that adjusts the dose intensity and direction from various directions. In this report, even compared to IMRT, the MLD, lung V5-20 Gy, and heart V20-40 Gy were significantly lower than those of the IMRT plan. Compared with IMRT, which allows for better dose adjustment than 3DCRT, PBT facilitates a good dose distribution.

A small irradiation field is a problem in the conventional irradiation system of PBT. Therefore, it is difficult to achieve a wide range of > 300 mm square, such as elective nodal irradiation (ENI)[16]. Although a patch-field irradiation technique in which multiple proton irradiation fields are connected is used, some institutions use X-ray therapy for ENI[43]. Sasaki et al[16] compared hybrid dose distribution using PBT and volumetric modulation arc therapy (VMAT), which is a rotating IMRT. When ENI was irradiated with VMAT and boost irradiation was performed with PBT, the dose reduction effect was smaller than that when both were irradiated with VMAT. The heart V30 Gy was reduced from 27.4% to 15.1%; however, the lung dose reduction effect was small. In contrast, when both were irradiated with PBT, the lung dose was also greatly reduced from 9.0 Gy to 5.9 Gy in MLD. Regarding the cardiac dose, studies have evaluated dose reduction in various parts of the heart, such as the left anterior descending artery (LAD). Shiraishi et al[19] reported that dose reduction was achieved with PBT compared with IMRT in the atrium, ventricle, and major coronary arteries, including the LAD. Regarding lung cancer, it has been pointed out that the radiation dose to the cardiovascular system, such as the LAD, may contribute to cardiotoxicity and, ultimately, survival prognosis[44]. In this report, V15 Gy of LAD ≥ 10% was associated with an increased risk of all-cause mortality (2-year OS was 47% vs 67%). Considering this threshold, Shiraishi et al[19] showed that even in the low-dose range of V5 Gy, the PBT group had a value of 8%, which was less than 10%. However, in the IMRT group, even in the high-dose range of V30 Gy, the value was 15.8%, which was significantly higher than 10%, making it extremely difficult to maintain the LAD threshold shown in a lung cancer study. Therefore, reducing the radiation dose to the vascular system, especially the LAD, would be more easily achieved using PBT than X-ray therapy.

In addition to the lungs and heart, analysis is being conducted on the effect of reducing bone marrow suppression. Warren et al[29] reported that only PBT plans showed significant sparing of the bone V10 Gy and bone mean dose, especially in patients with a larger irradiated field. This also means that the more advanced the cancer, the greater the significance of using PBT, as the irradiation area will be wider, even if the irradiation is focused on the cancer lesion without ENI.

COMPARING TOXICITIES BETWEEN PROTON BEAM AND X-RAY THERAPIES

As mentioned above, simulations show that radiation doses to the heart, lungs, and bones can be reduced using PBT compared to X-ray therapy; however, it is ultimately important to determine whether similar results can be achieved when treating actual patients in clinical situations. Table 1 summarizes the previous reports comparing the toxicities of PBT and X-ray therapies. Currently, there is only one phase 2 randomized controlled trial (RCT). Lin et al[30] reported the total toxicity burden (TTB), which is the cumulative severity of multiple toxicities in patients with EC after CRT with or without surgery. For example, grade 3 toxicities, including radiation pneumonitis, pleural effusion, and pericardial effusion weighed 60 points, whereas myocardial infarction weighed 70 points. In this report, the mean TTB of IMRT was 2.3 times higher than that of PBT, and the mean postoperative complication score of IMRT was 7.6 times higher than that of PBT. On the other hand, there have been only retrospective studies comparing IMRT and PBT in a few cases in which only definitive CRT was administered, and they reported no significant difference in toxicity between IMRT and PBT. However, the incidence rates of pneumonitis, pericardial effusion, and treatment-related deaths are lower in PBT[28]. Currently, most reports are retrospective comparisons, but a meta-analysis that included one RCT, as mentioned above, concluded that the incidence of grade 2 or higher pneumonitis and pericardial effusion was significantly lower in the PBT group compared to the X-ray therapy group with grade 2 or higher radiation pneumonitis was 2%, grade 2 or higher pleural effusion was 4%, grade 2 or higher pericardial effusion was 3%[31]. Therefore, PBT for EC can reduce cardiac and pulmonary toxicity, as demonstrated by its superior dose distribution in the simulation study. In addition, the benefit of a reduced dose is thought to be greater with a wider irradiation area, and PBT may offer the possibility of treatment in cases in which treatment would normally be hesitant owing to toxicity.

Table 1 Toxicities of previous reports which compared X-ray therapy and proton beam therapy (%).

Ref.	Modality	G ≥ 3 pericardial effusion	G ≥ 3 pneumonia	G ≥ 3 pleural effusion	G4 lymphocytopenia
Makishima et al[17]	3DCRT	0	5.3	5.3	-
	PBT	0	0	0	-
Shiraishi et al[19]	IMRT	-	-	-	40
	PBT	-	-	-	18
Choi et al[25]	IMRT	-	-	-	12.5
	PBT	-	-	-	20.0
Xi et al[28]	IMRT	2.4	2.9	1.9	-
	PBT	0.8	1.6	0.8	-
Lin et al[30]	IMRT	2	3	10	-
	PBT	0	0	2	-
Wang et al[32]	IMRT	-	-	-	52.5
	PBT	-	-	-	27.3
	PBT	0	-	0	40.7
Fang et al[33]	IMRT	-	-	-	60.5
	PBT	-	-	-	39.5
Routman et al[37]	XRT	-	-	-	60
	PBT	-	-	-	40
Zhu et al[38]	IMRT	-	-	-	46.2
	PBT	-	-	-	22.0

3DCRT: 3-dimetional conformal radiotherapy; G: Grade; IMRT: Intensity modulated radiotherapy; PBT: Proton beam therapy; XRT: X-ray therapy including both 3-dimetional conformal radiotherapy and intensity modulated radiotherapy.

There have been many comparative clinical reports on bone marrow suppression, particularly regarding lymphopenia[32-38]. Both the sub-analyses of the aforementioned RCT and meta-analysis reported that grade 4 or higher lymphopenia was significantly less likely to occur with PBT than with X-ray therapy, which coincides with the simulation results of the dose distribution advantage[31,32]. A sub-analysis of this RCT found that when baseline lymphocyte counts were less than 1.02 × 10³/μL, grade 4 or higher lymphopenia would occur in 100% of cases, regardless of whether IMRT or PBT was used. However, when the irradiation volume of the tumor was 358 mL or more and the baseline lymphocyte count was 1.95 × 10³/μL, there was a significantly large difference in the incidence of grade 4 or higher lymphopenia, with 35% for PBT and 70% for IMRT[32]. Therefore, the larger the irradiated field of the tumor, the greater the contribution of PBT to reducing lymphopenia, which is consistent with the results of the dose-volume analysis. In addition, according to a study using machine learning, reductions in grade 4 or higher lymphopenia risk were maximized with PBT in older patients with lower baseline absolute lymphocyte counts and higher lung and heart doses[38]. Although the volume of the irradiated tumor did not remain the most significant factor, it was surmised that high doses to the lungs and heart are more likely to occur when the tumor is large and a wide area must be irradiated. This report supports the results of the sub-analysis of phase 2 RCT. Furthermore, the effectiveness of PBT in elderly people, as shown in this study, suggests that PBT may play a larger role in an aging society in the future.

COMPARING CLINICAL RESULTS BETWEEN PROTON BEAM AND X-RAY THERAPIES

Currently, the effect of PBT on improving prognosis in the treatment of EC remains controversial. Clinical results are summarized for cases in which OS and progression-free survival (PFS) are reported in the text (Table 2). The phase 2 RCT showed no significant differences in OS or PFS[30]. However, several factors may have influenced the results of this phase 2 RCT, such as the fact that only half of the cases were assumed to be definitive settings, most underwent surgery and the PBT group had significantly more cases with a performance status of 1 or higher, despite being randomized. In addition, when evaluating the clinical results, it is important to consider that the recruitment of 67% of the patients was terminated because the discontinuation criteria for RCT on TTB were met. Therefore, there was insufficient statistical power to compare PFS with the other endpoints. Conversely, a retrospective comparative study conducted on several cases of definitive CRT reported that both OS and PFS were significantly better in the PBT group[28]. However, this report should be interpreted cautiously, as the median follow-up period was significantly lower in the PBT group (44.8 months vs 65.1 months). The meta-analysis also requires caution in interpretation as it includes many retrospective studies and cases of preoperative treatment and the use of data taken from graphs not specified in the text; however, similar to the retrospective comparison report mentioned above, the meta-analysis concluded that OS and PFS were significantly improved in the PBT group[31].

Table 2 Clinical results of previous reports in which clinical data were reported in the text.

Ref.	Modality	Definitive settings (%)	Number of patients	Median follow up (months)	OS (%)					PFS (%)
					1 year	1.5 years	2 years	3 years	5 years	1 years	2 years	3 years	5 years
Choi et al[25]	3DCRT	0	16	17.0	-	-	67.8	-	-	-	33.3	-	-
	PBT	0	15	17.0	-	-	68.6	-	-	-	34.5	-	-
Xi et al[28]	3DCRT	100	41	65.1	-	-		-	31.6	-	-	-	20.4
	PBT	100	132	44.8	-	-		-	41.6	-	-	-	34.9
Bhangoo et al[27]	IMRT	28	32	14	71	-	-	-	-	45	-	-	-
	PBT	19	32	10	74	-	-	-	-	71	-	-	-
Lin et al[30]	IMRT	50.8	61	44.1	-	-	-	50.8	-	-	-	44.8	-
	PBT	54.3	46	44.1	-	-	-	51.2	-	-	-	44.5	-
Sumiya et al[36]	3DCRT (all)	100	15	-	-	-	47.7	-	-	-	-	-	-
	PBT (all)	100	54	-	-	-	77.2	-	-		-	-	-
	3DCRT (≥ stage 2)	100	14	-	-	-	41.1	-	-	-	-	-	-
	PBT (≥ stage 2)	100	31	-	-	-	68.9	-	-		-	-	-
DeCesaris et al[40]	IMRT	0	36	-	-	59	-	-	-	-	-	-	-
	PBT	0	18	-	-	83	-	-	-	-	-	-	-

3DCRT: 3-dimetional conformal radiotherapy; IMRT: Intensity modulated radiotherapy; PBT: Proton beam therapy; PFS: Progression free survival; OS: Overall survival.

Conversely, while the numerical survival rate is important, survival with a maintained QOL is also important. There are few reports on long-term QOL after CRT, even after treatment with X-ray therapy[45]. The phase 2 RCT reported no significant difference in QOL between IMRT and PBT cases; however, the comparison was limited to 3 months or more at most, there was no annual comparison, and only a few people were evaluated. As mentioned in the Discussion section of this report, in addition to the lack of detection power for QOL[30], the inability to evaluate during the period when late toxicity occurs may have influenced this finding. Although it is difficult to properly evaluate long-term QOL, it is desirable to report on QOL evaluations annually. However, this phase 2 RCT showed a significant improvement in TTB considering toxicity in PBT cases, and that late toxicity may affect QOL, it is possible that performing PBT may improve QOL in the long term. If toxicity can be reduced by PBT, it may be possible to maintain survival while maintaining QOL compared with X-ray therapy.

Moreover, it has been suggested that lymphopenia may contribute to worsening prognosis not only in EC but also in other cancers[46-48]. Therefore, PBT may play a significant role in reducing the occurrence rate of severe lymphopenia.

COMPARING RESULTS BETWEEN PBT AND SURGERY

A direct comparison of a reasonable number of cases treated with preoperative chemotherapy followed by surgery and PBT with chemotherapy was recently reported in Japan[40]. In this study, the 3-year OS rates were 74.8% and 72.7% in the PBT and surgery groups, respectively, although the 3-year PFS was worse in the PBT group than in the surgery group (51.4% vs 59.6%, respectively). This result may be related to the significantly poorer prognosis after recurrence in patients who underwent surgery than those who underwent PBT (hazard ratio, 0.58). In particular, the results were better in the group that underwent salvage surgery or endoscopic treatment for local recurrence, which occurred more frequently than surgery. Although recurrence is common with PBT, careful follow-up after treatment may improve the prognosis significantly because 61% of the recurrences were local-only progressions in this report. Only one-third of the cases in this study underwent surgery after triplet chemotherapy, which is the current standard treatment in Japan. Although there is room for improvement in the treatment outcomes of the surgery group using triplet chemotherapy, this report suggests that it is possible to achieve treatment outcomes of PBT comparable to those of preoperative chemotherapy followed by surgery if appropriate treatment for recurrence after PBT is administered. In contrast, a similar report, although fewer cases, reported a worse prognosis in the PBT group for cT3-4 cases[49]. This result was thought to be influenced by the greater likelihood of local recurrence in patients undergoing PBT; therefore, we believe it is important to conduct follow-up observations assuming that therapeutic intervention will be performed for post-radiation recurrence. An RCT would be desirable to draw clear conclusions regarding the comparison between PBT and surgery. However, it is hoped that more rigorous prospective trials will be conducted, even though the treatments are completely different and difficult to perform.

Although no specific indicators were used in this study, Ogawa et al[41] speculated that preoperative chemotherapy followed by surgery might have a greater impact on the patient's general condition after treatment than PBT combined with chemotherapy. Although minimally invasive surgery improves QOL after surgery[50], extensive resection is unavoidable; therefore, surgery may result in a lower QOL, as was the case in a previous RCT[9] or meta-analysis[45] that compared X-ray therapy with surgery alone, which assessed QOL. In addition, 41.4% of the PBT group in this study underwent X-ray therapy during ENI; therefore, the difference in the impact on daily life may be even greater if treatment was performed with PBT alone. PBT has less toxicity than X-ray therapy; therefore, QOL may be maintained. Further investigation is needed in the future, with a particular focus on the differences in long-term QOL.

FUTURE

In the EC radiotherapy field, clinical trials of immune checkpoint inhibitor combination therapies are being conducted, as shown in Keynote 975[51]. It has been suggested that the efficacy of immune checkpoint inhibitors may be adversely affected by lymphopenia[52,53]. Colomb et al[52] reported that patients who undergo maintenance therapy with durvalumab after CRT for lung cancer may have a better prognosis if their lymphocyte counts are maintained. They concluded that maintaining lymphocyte counts is important. In addition, a meta-analysis examining lymphocyte count reduction due to all treatments and the effect of immune checkpoint inhibitors on lung cancer cases reported that lymphocyte count reduction contributed to a worsening prognosis[53]. This was a report on lung cancer, and since there are no current reports on EC, it is not known whether the same applies to EC; most of the reports were retrospective, but a similar effect may occur in EC. Therefore, the role of PBT in suppressing lymphopenia may become important if the Keynote 975 trial produces positive results.

In addition, future results, including a phase 3 RCT of PBT and IMRT with OS as the primary endpoint (NCT03801876) and a phase 1 clinical trial to confirm the effect of dose escalation with PBT in neoadjuvant CRT (NCT02213497), will likely clarify the role of PBT in EC compared to other treatment methods.

CONCLUSION

Although there is no conclusion regarding the OS improvement effect of PBT for EC compared with X-ray therapy, PBT reduces toxicity, such as cardiac, pulmonary, and bone marrow suppression. Therefore, even if OS does not improve, long-term QOL can be maintained in PBT cases. In addition, compared to radical surgery, if combined with appropriate recurrence treatment, good results can be obtained while suppressing the disadvantage of an increase in local recurrence.