Lymphopenia during chemoradiation—foe or friend

Background Severe lymphopenia during treatment is considered to be a poor prognostic factor. The current literature lacks information regarding its impact on various outcomes in locally advanced head-and-neck cancer patients in a prospective setting. Methods We recently published a randomised study comparing cisplatin–radiation with nimotuzumab cisplatin–radiation. The database of this study was used for the present analysis. The impact of severe lymphopenia (grade 4 lymphopenia) on progression-free survival (PFS), locoregional control (LRC) and overall survival (OS) was studied using the Kaplan–Meier method and Cox regression analysis. The binary logistic regression analysis was used to see the effect of various factors on the development of severe lymphopenia. Results We had a total of 536 patients, of which 521 patients (97.7%) developed lymphopenia. Grade 1 lymphopenia was noted in 10 (1.9%) patients, grade 2 in 100 (18.8%), grade 3 in 338 (63.1%) and grade 4 in 73 (13.7%) patients. The median PFS was 20.53 and 60.33 months in severe and non-severe lymphopenia, respectively (hazard ratio, 0.797; p-value = 0.208). The median duration of LRC was 56.3 months in severe lymphopenia, whereas it was not reached in non-severe lymphopenia (hazard ratio, 0.81; p-value = 0.337). The median OS was 28.46 versus 47.13 months in severe and non-severe lymphopenia, respectively (hazard ratio, 0.76; p-value = 0.11). Of various risk factors, gender was significantly associated with severe lymphopenia. Conclusion The occurrence of severe lymphopenia was not significantly associated with the outcomes. Gender is the only risk factor significantly linked to severe lymphopenia.


Introduction
Lymphocytes play a key role in tumour response; however, they are also the most sensitive cells to chemoradiation [1]. According to Campian et al [2], lymphopenia is commonly seen during the administration of chemoradiation to head-and-neck cancer (HNC) patients. In a prospective study, combined treatment with cisplatin and radiotherapy induced lymphopenia in 78% of patients [3]. In another phase I trial, lymphopenia (grade ≥ 3) was seen in 90% of patients receiving nimotuzumab along with chemoradiation for oesophageal cancer [4]. It has been postulated that lymphopenia hampers the body's ability to destroy tumour cells [5].
HNC shows the presence of many infiltrating immune cells including lymphocytes-hence, it is called an immunogenic tumour [6]. Furthermore, the pre-treatment lymphocyte counts can be of prognostic value in HNC as proposed by some previous studies [7]. Reportedly, low absolute numbers of CD3+, CD4+ and CD8+ T cells have been found in patients with head-and-neck squamous cell carcinoma in comparison to the normal controls. In addition, patients with active disease tend to show significantly lower CD3+ and CD4+ T cell counts than those with no disease [2].
The occurrence of lymphopenia and its impact on outcomes have never been evaluated in any prospective study in HNC patients. In view of this lacuna in the current literature, we performed this post hoc analysis to study the impact of severe lymphopenia on outcome (progressionfree survival, loco-regional control and overall survival) in HNC patients undergoing radical chemoradiation.

Materials and methods
Data set-We have recently published the results, outcome and adverse effect of a phase III randomised trial. In this study, 536 patients were randomised 1:1 to receive either radical radiotherapy (66-70 greys) with concurrent weekly cisplatin (30 mg/m 2 ) (CRT) or the same schedule of CRT with weekly nimotuzumab (200 mg) (NCRT). The primary endpoint was progression-free survival (PFS); key secondary endpoints were the duration of loco-regional control (LRC) and overall survival (OS) [8]. The database of this study was used for the present post hoc analysis.
Data collection-From the above-mentioned data set, the data of lymphopenia in addition to demography and outcome were extracted, and the following were noted: • The occurrence of lymphopenia from week 0 to week 6 of treatment was taken into account.
• The maximum grade of lymphopenia as per the CTCAE criteria 4.03.
Endpoints (outcomes)-To ascertain the impact of lymphopenia, the following endpoints were considered: • PFS was defined as duration from the date of randomisation to the first evidence of tumour progression. • LRC was defined as the time between the date of randomisation and the date of local or regional relapse. • OS was calculated as the time from the date of randomisation to the date of death from any cause.

Statistics
Statistical analysis was performed using the SPSS version 20 and R version 3.5.3. A descriptive analysis was performed. The continuous variables were expressed in median value along with its range, whereas the non-continuous/ordinal variables were expressed in percentage (%) along with 95% confidence interval (CI). The impact of severe lymphopenia on PFS, LRC and OS was studied using the Kaplan-Meier method and Cox regression analysis. The binary logistic regression analysis was used to see the impact of factors (age, gender, Eastern Cooperative Oncology Group performance score (ECOG PS) and chemotherapy regimen) on the development of severe lymphopenia. A p-value of 0.05 was considered to be significant.

Baseline characteristics
The detailed baseline characteristics of these patients have already been published [8]. The median age was 54 and 55 years in the CRT and NCRT arm, respectively. In the CRT arm, there were 86.2% of males and 13.8% of females, whereas, in NCRT arm, there were 84.3% of males and 15.7% of females. Stage IVA (64.2% versus 66%) was the most common stage, followed by stage III (32.5% versus 29.9%) and stage IVB (3.4% versus 4.1%) in CRT and NCRT, respectively. In both the arms, 99.6% of patients had ECOG PS of grade 0-1, whereas 0.4 % had grade 2. The baseline characteristics and treatment details of the patients are shown in Table 1.
Week-wise incidence and grade of lymphopenia are shown in Figure 1.
Apart from this, we also did an additional analysis to see any impact of lymphopenia on the outcomes. Here, we compared combined

Discussion
The existence of peripheral immune alterations in cancer patients was shown for the first time in the mid-1970s by Bone and Lauder [9]. Lymphopenia can be functional or treatment induced. Functional lymphopenia has been reported in advanced or metastatic breast cancer, colon cancer and hepatocellular carcinoma [7]. It is associated with dismal outcome. Furthermore, in various malignancies such as carcinoma breast, carcinoma ovary, non-Hodgkin lymphoma, diffuse large B-cell lymphoma, T-cell lymphomas, sarcomas and colon cancer, lymphopenia detected before treatment and early after starting chemotherapy (5-15 days) or radiotherapy is associated with a worse PFS and OS [7]. Its impact on relapse-free survival (RFS) and OS in various malignancies is shown in Table 3 [10][11][12][13][14][15]. In addition, the severity of lymphopenia has been reported as an independent prognostic factor of PFS and OS. According to Manuel et al [16], severe lymphopenia (<200/mm 3 ) in metastatic breast carcinoma has a strong impact on patient survival and also reported 40% and 90% reduction of the median OS or PFS, respectively. Head-and-neck squamous cell carcinoma (HNSCC) is known as an immunogenic tumour showing the presence of infiltrating lymphocytes and other immune cells [17]. The previous studies have suggested the important role of pre-treatment lymphocyte counts with regard to their association with prognosis in HNSCC [7]. However, Campian et al [2] in a retrospective study found no significant difference in PFS and OS between patients with TLC < 500 cells/mm 3 and those with TLC ≥ 500 cells/mm 3 . Similarly, in this study, there was no statistically significant impact of severe lymphopenia on various outcomes (PFS, LRC and OS).
Various studies have considered grade 3-4 as severe lymphopenia, and it was reported in 61% of HNC patients and 53% of cervical cancer patients by Campian et al [2] and Wu et al [18], respectively [18], whereas Zhou et al [17] found grade 4 lymphopenia in 31% of oesophageal cancer patients during chemoradiation. In this study, we found lymphopenia in 97.7% of patients, where, in 13.7% of patients, it was grade 4.
Older age, female gender, accelerated radiotherapy and certain chemotherapy agents (cisplatin, cyclophosphamide, methotrexate and taxanes) have been reported as significant risk factors for the development of lymphopenia [7,19]. In this analysis, we analysed these factors.
Of age, gender, ECOG performance score and chemotherapy regimen (cisplatin versus cisplatin + nimotuzumab), gender was the only significantly associated factor with severe lymphopenia. This first prospective study to enlighten lymphopenia and its impact on survival outcomes in head-and-neck cancer patients is unique. There are several limitations to this study. First of all, lymphopenia occurring only during treatment was analysed, so the post-therapy lymphopenia status cannot be commented on, although it is unlikely to witness post-treatment lymphopenia in a weekly chemotherapy schedule. Second, these data are applicable only for weekly cisplatin-based chemoradiation as there are different haematological toxicities in once a week versus once a 3-week cisplatin schedule [20].

Conclusion
Lymphopenia is fairly common in HNC patients during chemoradiation. Gender is an important factor associated with severe lymphopenia; however, there is no significant impact of severe lymphopenia on PFS, LRC and OS.