The impact of radiation-related lymphocyte recovery on the prognosis of locally advanced esophageal squamous cell carcinoma patients: a retrospective analysis

Abstract

Background

The impact of radiation-related lymphocyte recovery on prognosis in locally advanced esophageal squamous cell carcinoma (LA-ESCC) remains unclear.

Methods

Patients with stage II-IVa ESCC who received definitive RT were screened. Collect absolute lymphocyte counts (ALCs) before, during, and after RT. The recovery status of lymphocytes was observed at one-month post-RT (P1) and three months post-RT (P3). Patients with relatively lower lymphocyte recovery levels at P1 were divided into Group a and those with higher levels were divided into Group b. Patients with relatively lower lymphocyte recovery levels at P3 were divided into Group A and those higher were divided into Group B. Kaplan-Merier’s analysis and Cox analysis were conducted to compare survival outcomes. Binary logistic regression analyses was employed to ascertain factors associated with lymphocyte recovery.

Results

116 patients were enrolled. During RT, ALCs reached the bottom 5 weeks after RT started and 70.7% of patients experienced G3 lymphopenia. The median OS for Group a and Group b were 38.1 months and 14.4 months, p = 0.097. The median PFS for Group a and Group b were 14.2 months and 10.0 months, p = 0.037. Whereas, the median OS for Group A and Group B was 14.5 months and 22.2 months, p = 0.019. The median PFS for Group A and Group B were 8.4 months and 12.4 months, p = 0.021. Cox multivariable analysis revealed that higher lymphocyte recovery level at P3 was significantly associated with superior OS (p = 0.040, HR0.636) and PFS (p = 0.028, HR0.627). The logistic analysis identified a positive association between G4 lymphopenia during RT (p = 0.012, OR 7.742) and PTV dose < 60 Gy (p = 0.014, OR 2.655) with lymphocyte recovery.

Conclusions

The prognostic value of lymphocyte recovery status at P3 appears to be greater than that of lymphocyte recovery status at P1 for locally advanced ESCC patients. Radiation-related lymphocyte recovery might serve as a valuable prognostic factor in LA-ESCC.

Background

Esophageal carcinoma is a prevalent malignant tumor, with 455,800 new diagnoses and 400,200 deaths worldwide per year [1]. Radiotherapy (RT), a crucial anti-tumor approach, is a double-edged sword as it also depletes immune cells. Lymphocytes, which are vital in anti-tumor immune responses, are quite radiosensitive [2]. Preclinical and clinical studies showed that even low-dose irradiation (< 1 Gy) could kill circulating lymphocytes [3]. The fact that severe lymphopenia during radiotherapy is associated with inferior survival has been demonstrated in many solid tumors, including esophageal cancer [4,5,6]. Moreover, the response to immunotherapy is contingent upon peripheral absolute lymphocyte count. With the emergence of immunotherapy as a promising anti-cancer treatment, the significance of lymphocyte recovery from lymphopenia has become increasingly evident.

Many studies have been focusing on the impact of radiation-induced lymphopenia on the outcomes of esophageal cancer. Nevertheless, radiation-related lymphocyte recovery has received minimal attention. A study conducted in patients with locally advanced pancreatic cancer demonstrated a clear correlation between lymphocyte recovery within 6 months of initiating chemoradiotherapy (CRT) and superior clinical outcomes [7]. Another study found recovery from lymphopenia within 3 months after concurrent CRT ended is significantly correlated with improved survival among non-small cell lung cancer (NSCLC) patients undergoing concurrent CRT and adjuvant immunotherapy [8]. The significance of lymphocyte recovery in esophageal cancer remains a topic of debate. Tseng et al. reported that inadequate lymphocyte recovery at 6 months following definitive CRT was associated with unfavorable outcomes in patients with locally advanced esophageal squamous cell carcinoma (ESCC) [9]. In contrast, a separate study conducted by Deng et al. indicated that lymphocyte recovery at 6–8 weeks following CRT appears to be unrelated to long-term outcomes in esophageal cancer [10]. The prognostic value of radiation-related lymphocyte recovery remains a topic of contention, and the optimal time point for defining lymphocyte recovery remains to be elucidated.

Building upon previous research findings, this study aimed to explore the relationship between radiation-related lymphocyte recovery and survival outcomes in patients with locally advanced ESCC. Additionally, the study sought to identify an optimal time point for observing lymphocyte recovery status.

Methods

Patients

Patients with locally advanced ESCC underwent definitive radiotherapy between January 2015 and December 2021 were screened.

Inclusion criteria: (1) Age ≥ 18, (2) Eastern Cooperative Oncology Group (ECOG) score ≤ 2, (3) Pathologic confirmation of ESCC, (4) Stage II-IVa, unresectable disease (e.g. cervical oesophageal cancer, difficult to achieve R0 resection, contraindications to surgery or high risk of surgery, etc.) or patient refusing to undergo surgery. (5) Patients who have undergone initial definitive radiotherapy or definitive chemoradiotherapy, (6) Radiation dose of at least 50 Gy, (7) Absence of prior or concomitant malignancy, (8) Complete hematology tests data. Exclusion criteria: (1) Patients with diseases that might affect lymphocyte count (e.g., hematologic malignancies, severe infection, or immunosuppression not caused by chemoradiotherapy), (2) Patients undergone surgery, (3) Interruption of radiotherapy for more than 3 consecutive days, (4) Combination immunotherapy, (5) Pregnancy, (6) Lack of clinical or follow-up information.

Patients’ clinical information is derived from the electronic medical record of our institution. All patients signed an informed consent form for treatment. This retrospective analysis was approved by the appropriate institutional review board.

Treatment and follow-up

All patients received a total dose of 50.0–66.0 Gy (5 days/week, 1.8–2.0 Gy/d), using intensity-modulated radiation therapy (IMRT) with involved-field irradiation (IFI), with or without chemotherapy. The chemotherapy regimens were generally doublets, comprising a paclitaxel, fluorouracil, or platinum-based compound (TF, PF, or TP). Complete blood counts (CBCs) were obtained at the following time points: before the initiation of RT, weekly throughout RT, at one-month post-RT, and three months post-RT. Patients were followed up every 3 months for the first two years after the whole treatment and every 6 months to the fifth year, then annually until death.

Data collection

Absolute lymphocyte count (ALC) values were collected before RT (Pre-RT), every week during RT (W1-W5), at the end of RT (Post-RT), at 1 month post-RT (P1), and 3 months post-RT (P3). ΔALC₁ was defined as ALC P1 minus ALC Post-RT. ΔALC was defined as ALC P3 minus ALC Post-Rt.

The lowest ALC during RT was identified as the minALC and was graded according to the Common Terminology Criteria for Adverse Events (CTCAE) version 5.0.

Statistical analyses

Patients were grouped according to ΔALC₁ or ΔALC. The optimal cut-off points of ΔALC₁ and ΔALC were determined by the maximally selected log-rank statistics based on OS using the R package ‘maxstat’. The clinical baseline data of different groups were compared using the Chi-square test. Endpoints were overall survival (OS) and progression-free survival (PFS), which were defined from the date of treatment start until the events or censoring. Kaplan–Meier’s analysis and log-rank test were employed to compare survival outcomes (OS and PFS) between different groups. Multivariable Cox proportional hazard regression analyses were used to ascertain the factors associated with PFS and OS. A binary logistic regression model was constructed to examine the potential association between lymphocyte recovery and various factors. All statistical tests were two-sided; p values < 0.05 were considered statistically significant. Chi-square test, Cox proportional hazard regression analyses, and binary logistic regression analysis were conducted using SPSS, version 27.0. Kaplan–Meier’s analysis and maximally selected log-rank statistics were done by R software 4.3.2.

Results

Patient characteristics

A total of 116 patients were included (Table 1). Most patients were male (61.2%). Approximately half (53.4%) of the patients had an ECOG performance status score of 0, with a median age of 67 years. The median tumor length was 5 cm, with 15 cases (12.9%) located in the cervical esophagus, 30 cases (25.9%) located in the upper thoracic esophagus, 36 cases (31.0%) located in the middle thoracic esophagus, and 35 cases (30.2%) were located in the lower thoracic esophagus. 32 patients (27.6%) had stage II disease, 46 (39.7%) had stage III, and 38 (32.8%) had stage IVa. 89 patients (76.7%) received definitive chemoradiotherapy, and 27 (23.3%) patients received definitive radiotherapy. Most of the chemotherapy regimens were TP, FP, and TF. The duration of the chemotherapy cycle ranged from one to six cycles. Clinical characteristics in detail are listed in Table 1.

Table 1 Baseline clinical characteristics

Survival of total patients

At the last follow-up date October 31, 2023, the median follow-up time was 41 months. The median PFS was 10.2 months. 1-year, 2-year and 3-year PFS was 45.7%, 25.7% and 23.7%, respectively. The median OS of 116 patients was 17 months. 1-year, 2-year and 3-year OS was 75.0%, 38.6% and 32.7%, respectively.

ALCs during and after RT

During treatment, the ALCs exhibited a general decline, reaching a nadir at W5. After RT finished, ALC gradually elevated to near-normal levels. The median ALC before RT, during RT, and at the end of RT were (×10⁹/L) 1.74, 1.13, 0,81, 0.60, 0.49, 0.41, 0.42, respectively (Fig. 1). During RT, 3 (2.6%) patients experienced grade 1 (G1) lymphopenia, 19 (16.4%) patients experienced G2, 82 (70.7%) patients experienced G3, and 12 (10.3%) patients experienced G4.

Fig. 1

ALCs during and after RT. Note: pre: pre-RT, w1: week 1, w2: week 2, w3: week 3, w4: week 4, w5: week 5, post: post-RT, p1: 1 month after RT finished, p3: 3 months after RT finished

Median P1 ALC was 0.97 × 10⁹/L, and median P3 ALC was 0.86 × 10⁹/L. 36 (31.0%) of the 116 patients had ALC recovered to normal levels, while 31 (26.7%) still had G1 lymphopenia, 38 (32.8%) had G2, 11 (9.5%) had G3, and no patient had grade 4 lymphopenia 3 months after RT completion.

The prognosis predictive value of ΔALC1 and ΔALC

The optimal cut-off point of ΔALC₁ was 0.10. Patients were divided into Group a (ΔALC₁ ≤ 0.10) and Group b (ΔALC₁ > 0.10) based on this cut-off point. The median OS of Group a and Group b were 38.1 months VS. 14.4 months, p = 0.097. The median PFS of Group a and Group b were 14.2 months VS. 10.0 months, p = 0.037. The optimal cut-off point of ΔALC was 0.42. Patients were divided into Group A (ΔALC ≤ 0.42) and Group B (ΔALC > 0.42) according to this cut-off point. The median OS was 14.5 months in Group A and 22.2 months in Group B, p = 0.019. The median PFS was 8.4 months in Group A and 12.4 months in Group B, p = 0.021.

When patients were grouped according to lymphocyte recovery value at 1-month post-RT (ΔALC₁), no significant difference was observed in OS between the two groups. Conversely, both OS and PFS exhibited significant differences between the 2 groups when grouping patients by lymphocyte recovery value at 3 months post-RT (ΔALC). Consequently, the subsequent analysis will utilize the lymphocyte recovery value at 3 months post-RT.

The baseline characteristics of Group A and Group B are in Table 1.

Survival comparison between Group A and Group B

The median OS was 14.5 months in Group A and 22.2 months in Group B, p = 0.019. The 1-year, 2-year, and 3-year OS in Group A was 61.7%, 27.9%, and 23.6%, respectively. The 1-year, 2-year, and 3-year OS in Group B was 89.3%, 50.0%, and 42.2%, respectively (Fig. 2A). The median PFS was 8.4 months in Group A and 12.4 months in Group B, p = 0.021. The 1-year, 2-year, and 3-year PFS in Group A were 38.3%, 14.4% and 14.4%, respectively. The 1-year, 2-year, and 3-year PFS Group B were 53.6%, 37.5%, and 33.3%, respectively (Fig. 2B).

Fig. 2

Kaplan-Meier curves of OS (A) and PFS (B) between Group A (ΔALC ≤ 0.42) and Group B (ΔALC > 0.42)

Subgroup analysis

The results of subgroup analyses are shown in Table 2.

Table 2 Prognostic value of ΔALC in different subgroups

Factors related to lymphocyte recovery

Univariable logistic regression analysis showed that G4 lymphopenia during RT (p = 0.021, OR 6.304, 95%CI 1.316–30.204) and PTV dose < 60 Gy (p = 0.029, OR 2.303, 95%CI 1.090–2.864) were predictors of higher ΔALC. In multivariable logistic regression analysis, G4 lymphopenia during RT (p = 0.012, OR 7.742, 95%CI 1.560–38.410) and PTV dose < 60 Gy (p = 0.014, OR 2.655, 95%CI 1.216–5.794) was positively associated with higher ΔALC (Table 3).

Table 3 Factors associated with lymphocyte recovery

Other factors related to survival

Multivariable COX regression analysis showed that T < 3 (p < 0.001, HR: 0.421, 95%CI: 0.266–0.666) and higher ΔALC (p = 0.040, HR: 0.636, 95%CI: 0.413–0.979) were independent factors associated with superior OS. T < 3 (p < 0.001, HR: 0.445, 95%CI: 0.282–0.705), N < 2 (p = 0.008, HR: 0.539, 95%CI: 0.341–0.852), combined chemotherapy (p = 0.023, HR: 0.564, 95%CI: 0.344–0.923) and higher ΔALC (p = 0.028, HR: 0.627, 95%CI: 0.413–0.950) were independent factors associated with superior PFS (Table 4).

Table 4 Univariate and multivariate analysis for survival outcomes

Discussion

This study indicates that radiation-related lymphocyte recovery at 3 months post-RT was associated with superior survival in patients with locally advanced ESCC who have undergone definitive RT.

In this study, we graphed the weekly trend of ALC. Similar to other studies [7,8,9,10], we observed that lymphocytes declined rapidly within 3 weeks of the initiation of RT, subsequently declining gently until they reached a nadir at W5. Subsequently, ALC remained stable until the conclusion of RT, which indicated that ALC might reach a plateau following irradiation at a dose of 50 Gy. A meta-analysis revealed that, in comparison to a radiation dose of approximately 50 Gy, a dose of approximately 60 Gy was associated with improved local control in esophageal cancer patients. However, this did not result in a reduction in distant metastasis or an improvement in survival [11]. This phenomenon may be attributed to the fact that the ALCs had already reached a nadir following irradiation with 50 Gy. As the radiation dose increased to 60 Gy, the number and functionality of lymphocytes could not be restored, resulting in impaired immune surveillance evasion, facilitating the development of distant metastases and a poor prognosis. Moreover, 60 Gy was associated with an increased incidence of adverse events. Consequently, it is not advisable to increase the RT dose from 50 Gy to 60 Gy.

The prognostic value of radiation-induced lymphopenia (RIL) has been well studied. Many researchers have found that severe lymphopenia during RT predicted worse outcomes in esophageal cancer patients [12, 13]. However, we did not find that G4 ALC nadir was related to worse survival. This discrepancy may be attributed to the transient nature of the ALC nadir during RT, which may not fully reflect the anti-tumor immunity. In contrast, the dynamic changes of ALCs after RT, which is lymphocyte recovery after RT, represent a more reliable biomarker for patients’ anti-tumor immunity and prognosis. Some patients may experience severe RIL, but they may have a longer survival if they have a higher level of lymphocyte recovery status after RT has been completed.

The field of lymphocyte recovery has seen a surge in interest in recent years. Lee et al. reported that lymphocyte recovery (defined as ALC ≥ 0.8 × 10⁹/L 6 months after RT) was associated with longer OS and PFS in locally advanced pancreatic cancer patients who received concurrent CRT [7]. Cho et al. found that lymphocyte recovery (defined as ALC ≥ 0.5 × 10⁹/L 3 months after CRT) predicts superior survival of patients undergoing concurrent CRT and adjuvant immunotherapy [8].

However, the prognostic value of radiation-related lymphocyte recovery in esophageal cancer remains a topic of contention. Ihsuan et al. demonstrated that the ratio of ALCs at 6 months after definitive CCRT to ALCs at baseline (lymphocyte recovery index, LRI) ≥ 60% predicted superior survival in ESCC patients [9]. In contrast, Deng, et al. reported that lymphocyte recovery (defined as ALC at 6–8 weeks after CRT ≥ 0.8 × 10⁹/L) did not mitigate the negative impact on survival outcomes caused by radiation-induced G4 ALC nadir [10]. Two potential explanations exist for the findings of Deng’s study. Firstly, 81.8% of the study population had esophageal adenocarcinoma. Consequently, the pattern that Deng found might not apply to patients with esophageal squamous cell carcinoma, which represents the majority in Asia. Secondly, they defined lymphocyte recovery as ALC reaching 0.8 × 10⁹/L 6–8 weeks after RT ended. This might not be sufficient time for lymphocyte counts to recover. Ihsuan et al. defined lymphocyte recovery as LRI ≥ 60% at 6 months after the end of CRT, which is longer than previous studies. Their findings demonstrated its prognostic value in ESCC patients.

There are still debates on the time point defining lymphocyte recovery after RT. In our study, the status of lymphocyte recovery was observed at two time points: 1 month after RT finished and 3 months after RT finished. Interestingly, relatively lower lymphocyte recovery status at P1 was related to better survival, which was in contrast to the results of lymphocyte recovery status at P3. This may be attributed to the fact that the majority of patients did not complete their chemotherapy regimen at 1 month post-RT, resulting in lymphocyte recovery status affected by multiple factors and were unstable. However, the total patients had all treatment completed at 3 months post-RT. Previous studies also indicated that ALC remained stable 3 months after CRT started [7]. Moreover, statistics showed that lymphocyte recovery status at 3 months post-RT was a better biomarker of prognosis than lymphocyte recovery status at 1-month post-RT. Consequently, we elected to further investigate the impact of lymphocyte recovery status at three months post-RT on prognosis.

There is currently no consistent conclusion on when should we focus on lymphocyte recovery. Studies investigating the prognostic value of lymphocyte recovery after RT are limited. In esophageal cancer, lymphocyte recovery at 6–8 weeks after CRT did not appear to be associated with long-term outcomes in esophageal cancer [10], whereas lymphocyte recovery at 6 months after definitive CRT was associated with favorable outcomes [9]. Studies in other cancers have also debated the best time to focus on lymphocyte recovery. Lymphocyte recovery within 6 months of starting CRT appeared to be associated with better clinical outcomes in patients with locally advanced pancreatic cancer [7]. Recovery from lymphopenia within 3 months of CRT was correlated with improved survival in patients with non-small cell lung cancer [8]. In conclusion, the best time to define lymphocyte recovery remains to be explored. Overall, 3–6 months after completion of treatment may be appropriate. Less than 3 months may not be sufficient for lymphocyte recovery, and more than 6 months may be too late for follow-up. Large studies are needed to determine the optimal time point for lymphocyte recovery.

In the subgroup analysis, we found that lymphocyte recovery seemed to be associated with survival outcomes in patients who received definitive RT, but not in those who received definitive CRT. On the one hand, the limited sample size might contribute to the results that lymphocyte recovery was not associated with PFS and OS in patients who underwent CRT. On the other hand, CRT might cause more severe myelosuppressive effects than RT alone, which means that 3 months might not be enough for patients who underwent CRT to recover from lymphopenia. The best time to observe lymphocyte recovery after definitive CRT may be longer than 3 months after treatment, which is consistent with a previous study [9]. Further large-scale studies should be conducted to investigate the prognostic value of lymphocyte recovery after definitive CRT in ESCC patients and the best time point to observe lymphocyte recovery.

In multivariate analysis, we found that patients who underwent definitive CRT had better treatment outcomes, which was in line with previous study [14]. On the one hand, chemotherapy killed tumor cells directly. It could also act as a promoter of the local-regional effects of RT. On the other hand, chemotherapy eradicated some subclinical distant metastases, thus contributing to longer PFS.

Interestingly, logistic regression analyses showed that G4 lymphopenia during RT was positively associated with higher ΔALC, which is inconsistent with other similar studies [7–8]. However, the above studies defined lymphocyte recovery as ALC reaching a specific absolute level after a period of time following RT, so ALCs of patients who experienced a lower ALC nadir during treatment may be less likely to recover to the specific level. Instead, our study defined lymphocyte recovery as a dynamic difference between the ALC at 3 months post-RT and the ALC at the end of RT. Some patients who experienced G4 lymphopenia during RT may also recover faster than other patients after RT and have a better prognosis.

If the patient’s lymphocyte recovery is not ideal, what kind of treatment strategy should be used? Preclinical studies have provided some insight. For example, IL-7 has been shown to alleviate radiation-induced lymphopenia, promote lymphocyte infiltration into tumors, and improve outcomes in combination with radiotherapy [15–16]. Amifostine has also been shown to enhance the recovery of CD95 + T cells and NK cells in head and neck cancer patients undergoing RT, and may enhance the efficacy of the latter by interfering with FAS-related immunological pathways [17]. Unfortunately, there are currently no clinical trials that investigate strategies for promoting lymphocyte recovery after RT. What we can do is keep an eye on patients’ ALC levels after RT, monitor them regularly, and refer patients for regular imaging.

Our study has limitations. First, a small proportion of patients did not receive chemotherapy, which introduces multiple confounding factors into the analysis. Further large-scale studies are needed to investigate the effect of chemotherapy on radiation-induced lymphocyte recovery. Second, this is a retrospective study, which means that we could not take into account all factors that might influence lymphocyte counts during RT. Third, considering the limited sample size, the cut-off value of ΔALC (0.42) may need to be validated by external validation or large sample studies before being generally applied to other datasets to accurately predict the prognosis of LA-ESCC patients undergoing RT. Fourth, flow cytometry was not used to identify lymphocyte subtypes, which play a key role in the immune response. Further testing of peripheral blood immune cell subtypes may allow clinicians to select patients who would benefit from RT + immunotherapy and predict prognosis with greater accuracy. Finally, data on dose to organs at risk of lymphopenia (heart, large vessels, lung, mediastinal lymph node drainage area and bone marrow, etc.) should be collected and their relationship with radiation-induced lymphopenia and lymphocyte recovery should be analyzed.

Despite these limitations, our study suggests that radiation-related lymphocyte recovery is associated with improved survival in locally advanced ESCC patients. These findings need to be confirmed in prospective studies. Promoting lymphocyte recovery after RT may be a potential avenue for future therapeutic applications.

Conclusion

Lymphocyte recovery status at 3 months post-RT has more prognostic value than lymphocyte recovery status at 1-month post-RT in locally advanced ESCC patients. Radiation-related lymphocyte recovery is associated with superior OS and PFS in locally advanced ESCC.