Correlation between pelvic bone marrow dose volume parameters and acute hematologic toxicity in neoadjuvant intensity-modulated radiotherapy for locally advanced rectal cancer

doi:10.21203/rs.3.rs-4813077/v1

Download PDF

Research Article

Correlation between pelvic bone marrow dose volume parameters and acute hematologic toxicity in neoadjuvant intensity-modulated radiotherapy for locally advanced rectal cancer

https://doi.org/10.21203/rs.3.rs-4813077/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Objective

To explore the correlation between pelvic bone marrow dose volume parameters (Vx) and acute haematological toxicity (HT) in patients undergoing neoadjuvant intensity- modulated radiotherapy for locally advanced rectal cancer(LARC), and to provide clinical data to reduce the risk of radiotherapy-associated HT and to optimise radiotherapy planning.

Methods

Seventy-five patients with LARC who underwent neoadjuvant concurrent radiotherapy (CCRT) from October 2019 to November 2023 at the Affiliated Cancer Hospital of Guizhou Medical University were retrospectively analysed. All patients underwent 5-field intensity- modulated radiotherapy (IMRT) with a prescribed dose of 45-50.4 Gy/25–28 times in the planned target (PTV). Simultaneous capecitabine or 5-fluorouracil (5-Fu) chemotherapy was delivered during radiotherapy. Acute HT during radiotherapy was assessed according to the National Cancer Institute Common Toxicity Criteria Version 5.0 (NCI-CTC.V5.0). Logistic regression analyses were used to assess the relationship between the dose-volume of bone marrow in the pelvis (Vx of hip, sacrum, and femur) and acute HT, and the non-linear relationship and threshold effect between the two were further analysed using the generalised additive model (GAM) and segmental regression.

Results

Multifactorial logistic regression analysis showed that there was a significant correlation between total pelvic bone marrow irradiated volume (TV) and coxal bone marrow irradiated volume (CV) at low doses (V₅, V₁₀) and ≥ 2 degrees of leukocyte reduction and neutrophil reduction (P < 0.05).There was a significant negative correlation between low-dose (V₅, V₁₀, V₁₅) sacral bone marrow irradiated volume (SV) and ≥ 2 degrees of leukocyte reduction (P < 0.05). Threshold effect analysis showed that there was a threshold effect of CV₁₀ with both ≥ 2 degrees of leukopenia and neutropenia, with similar thresholds of 483–495 ml for both.

Conclusion

In neoadjuvant intensity-modulated radiotherapy for rectal cancer, CV is a better predictor of acute HT compared with TV. CV associated with acute HT is mainly concentrated at low dose levels (CV₅, CV₁₀, CV₁₅). The threshold of 483 ml for CV₁₀ can be used as a reference for reducing the risk of acute HT and optimising radiotherapy planning in pelvic radiotherapy.

Rectal cancer

Neoadjuvant radiotherapy

Intensity-modulated radiotherapy

Dose-volume parameters

haematological toxicity.

Adjuvant chemoradiotherapy improves local control and overall survival in locally advanced rectal cancer (LARC) (1, 2), but neoadjuvant chemoradiotherapy can further improve local control and pathologic complete response (pCR) (3–5). Therefore, neoadjuvant chemoradiotherapy has become the standard treatment modality for LARC. When LARC patients undergo neoadjuvant concurrent chemoradiotherapy (CCRT), the pelvic bone marrow will inevitably be irradiated at a certain dose, which may lead to different degrees of acute hematological toxicity (HT).Previous studies have shown that in concurrent radiotherapy combined with single-agent fluorouracil, the incidence of acute HT can reach 6%-20% (2, 3, 6, 7), and radiotherapy combined with stronger concurrent chemotherapeutic regimens can reach 25%-30% of grade 3–4 acute HT while achieving higher pCR rates(6, 8).Frequent or severe acute HT increases the risk of infection and hospitalisation costs, reduces treatment tolerance and even decreases treatment efficacy(9, 10).Several previous studies have shown that acute HT correlates with pelvic bone marrow (PBM) dose volume in pelvic radiotherapy (11–16), but there is little consistency in conclusions across studies and fewer studies have addressed HT in CCRT for rectal cancer.Therefore, further exploration of the true relationship between acute HT due to CCRT and PBM dose volume in LARC is particularly important for controlling acute HT. The aim of this study was to investigate the relationship between PBM dose volume and acute HT in LARC patients undergoing CCRT and to identify valid predictors and thresholds to provide a clinical basis for optimising radiotherapy planning and reducing acute HT.

1.Clinical data:

Data of 80 LARC patients treated with neoadjuvant CCRT from October 2017 to November 2023 at the Affiliated Cancer Hospital of Guizhou Medical University were retrospectively collected. Inclusion creteria: KPS ≥ 80; pathologically confirmed rectal adenocarcinoma; no obvious contraindication to radiotherapy. Exclusion criteria: previous pelvic radiotherapy; history of other tumours.Among the 80 patients, 2 cases had leukocytes lower than standard before treatment, and 3 cases did not complete CCRT due to personal reasons, so the final 75 patients were included in the analysis.

2.Treatment plan:

Radiotherapy was performed with 5-field intensity- modulated radiotherapy (IMRT) using 6 MV X-rays in the prone position (Swedish Elekta linear accelerator). All target areas and organs at risk (OARs) were outlined and dose limits for OARs were performed according to Radiation Therapy Oncology Group 0822 (RTOG 0822). All patients were treated with concurrent 5-FU (225 mg/m2-d) or capecitabine (875 mg/m2, 2 times/d) chemotherapy during radiotherapy.

3.Bone marrow region outlining and dose-volume parameters:

The pelvis was outlined on the bone window of the Pinnecal 3.0 system, and the outlining was based on the study of Yang et al (9). The PBM was divided into 3 anatomical sites, as shown in Fig. 1:(i)Coxal Bone Marrow (CBM): the bone marrow of the hip bone, sciatic bone and pubic bone bilaterally. (ii) Sacral bone marrow (SBM): the bone marrow extending from the L5 ~ S1 junction to the coccyx. (iii) Femoral bone marrow (FBM): bone marrow of the femoral head and proximal femur bilaterally. Coxal bone marrow volume (CV), sacral bone marrow volume (SV) and, femoral bone marrow volume (FV) were obtained from the intra-systemic dose-volume histogram (DVH). The total pelvic bone marrow volume (TV) is the sum of the three. The bone marrow volume parameters that received at least x Gy irradiation were Vx (V₅, V₁₀, V₁₅, V₂₀, V₂₅, V₃₀, V₃₅, V₄₀).

Orange is the sacrum; cyan is the coxal bone; blue and purple are the right and left proximal femur, respectively.

Figure 1: Extent of sacral, coxal bone, and femoral outlining

4.haematological toxicity assessment: Patients' pre-treatment complete blood counts were within normal limits. At least once a week from the start of CCRT to 3 weeks after the end of radiotherapy, patients underwent a complete blood count to record the most severe toxic events including white blood cells (WBC), neutrophils (ANC), haemoglobin (HB), and platelet count (PLT). Acute HT was assessed and graded by the National Cancer Institute Common Toxicity Criteria Version 5.0 (NCI-CTC.V5.0).

Statistical Analysis

The research variables in this study were patient total pelvis (TV) and different pelvic regions (CV, SV, FV) dose volume parameters (Vx).The outcome variables were all types of acute HT of grade ≥ 2. The outcome variables were all types of acute HT of grade ≥ 2. Relevant covariates included patient age, gender, body mass index (BMI), cT stage, cN stage, and concurrent chemotherapy regimen. Correlations between bone marrow irradiated volume at different dose levels and different degrees of acute HT were analysed using univariate and multivariate logistic regression. Generalised additive model (GAM) and segmental regression were applied to detect potential non-linear relationships and threshold effects between PBM dose volume and various types of acute HT. Data analysis was done using Epower 2.2 (X&Y Solutions, USA).P < 0.05 was considered as statistically significant difference.

1.general information: The number of patients in the study who occurred ≥2 degrees of leukopenia, neutropenia, haemoglobinopenia, and plateletopenia were 34, 25, 8, and 2, respectively, with a percentage of the total number of patients in the study of 45.33%, 33.33%, 10.67%, and 2.67%. The distribution of relevant research variables in acute HT is shown in Table 1.The mean value of age was 57.5 years, the mean value of BMI was 21.8 kg/m2, the mean value of TV was 1354 ml, and the difference in the distribution of each clinical variable was not statistically significant.

Table 1 The distribution of related clinical factors

	leukopenia						neutropenia
	＜2 degree		≥2 degree		P-value	Test value	＜2 degree			≥2 degree		P-value	Test value
	cases	%	cases	%			cases		%	cases	%
gender					0.203	1.620						0.851	0.035
male	33	44.0%	23	30.7%			37	49.3%		19	25.4%
female	8	10.7%	11	14.6%			13	17.3%		6	8.0%
Chemotherapy regimen					0.251	1.318						0.225	1.470
5-Fu	25	33.3%	25	33.3%			31	41.3%		19	25.4%
Cape	16	21.4%	9	12.0%			19	25.3%		6	8.0%
Tumor location					0.307	0.858						0.263	0.877
Lower	11	14.7%	9	11.9%			12	16.0%		7	9.3%
Middle	19	25.3%	14	18.7%			23	30.7%		10	13.3%
Upper	11	14.7%	11	14.7%			15	20.0%		8	10.7%
T Stage					0.658	-*						0.909	-*
2	2	2.7%	1	1.3%			2	2.7%		1	1.3%
3	21	28.0%	21	28.0%			27	36.0%		15	20.0%
4	18	24.0%	12	16.0%			21	28.0%		9	12.0%
N Stage					0.429	1.692						0.693	0.734
0	7	9.3%	10	13.3%			11	14.7%		6	8.0%
1	14	18.7%	9	12.0%			14	18.7%		9	12.0%
2	20	26.7%	15	20.0%			25	33.3%		10	13.3%

Abbreviations: BMI is body mass index, TV is total pelvic bone marrow volume, Cape is capecitabine, and * indicates the use of fisher's exact probability method.

2.Logistic regression analysis: univariate and multivariate logistic regression analysis model I (covariates: age, gender, cT stage, cN stage, concurrent chemotherapy regimen) showed that TV₅, CV₅-CV₁₅ were significantly associated with ≥2 grade leukopenia, and TV₅-TV₁₀, CV₅-CV₁₅ were significantly associated with ≥2 grade neutropenia (Tables 2-5). Model II (covariates: age, sex, cT stage, cN stage, concurrent chemotherapy regimen, same dose of SV) showed that TV₅-TV₁₀ and CV₅-CV₂₀ were significantly associated with ≥2 grade leukopenia, TV₅-TV₁₀ and CV₅-CV₁₅ were significantly associated with ≥2 grade neutropenia.In the multifactorial logistic regression analysis of SV and acute HT, SV₅-SV₁₅ were significantly associated with ≥2 grade leukopenia only in model II (adding the same dose level of CV to the covariates), and the OR values were all less than 1.00. No significant correlation was seen between any of the FV_X and acute HT.

Table 2 Sensitivity analysis of ≥2 grade leukopenia and neutropenia related to each irradiated TV_X based on Logistic regression

		≥2 grade leukopenia			≥2 grade neutropenia
	model	OR	95%CI	P-value	OR	95%CI	P-value
TV₅	Not adjusted	1.02	0.99，1.04	0.225	1.04	1.01,1.07	0.022
	model I	1.04	1.00，1.09	0.003	1.06	1.02,1.11	0.008
	model II	1.14	1.05，1.37	<0.001	1.10	1.02,1.18	0.010
TV₁₀	Not adjusted	1.01	0.98，1.04	0.463	1.03	1.00,1.06	0.056
	model I	1.03	0.99，1.07	0.104	1.05	1.01,1.10	0.018
	model II	1.11	1.03，1.20	0.006	1.08	1.00,1.17	0.041
TV₁₅	Not adjusted	1.00	0.98，1.03	0.828	1.02	0.99,1.05	0.119
	model I	1.02	0.98，1.06	0.335	1.04	1.00,1.09	0.052
	model II	1.06	1.00，1.12	0.066	1.06	0.99,1.13	0.098
TV₂₀	Not adjusted	1.00	0.97，1.02	0.876	1.02	0.99,1.05	0.184
	model I	1.01	0.97，1.04	0.724	1.03	0.99,1.07	0.091
	model II	1.02	0.97，1.07	0.456	1.03	0.97,1.09	0.342
TV₂₅	Not adjusted	0.99	0.96，1.02	0.435	1.01	0.98,1.04	0.544
	model I	0.99	0.96，1.02	0.581	1.01	0.98,1.05	0.466
	model II	0.99	0.95，1.04	0.756	0.99	0.95,1.04	0.829
TV₃₀	Not adjusted	0.98	0.95，1.02	0.323	1.00	0.97,1.04	0.793
	model I	0.98	0.94，1.02	0.355	1.00	0.96,1.04	0.849
	model II	0.99	0.93，1.05	0.748	0.98	0.93,1.04	0.455
TV₃₅	Not adjusted	0.98	0.94，1.02	0.396	1.00	0.96,1.05	0.846
	model I	0.97	0.92，1.03	0.338	1.00	0.95,1.05	0.968
	model II	0.99	0.90，1.08	0.822	0.95	0.88,1.04	0.285
TV₄₀	Not adjusted	0.99	0.94，1.04	0.614	1.01	0.95,1.06	0.829
	model I	0.98	0.92，1.05	0.549	1.00	0.94,1.07	0.969
	model II	1.03	0.90，1.17	0.708	0.94	0.84,1.07	0.362

The exposure variable is the total pelvic bone marrow volume (TV₅-TV₄₀), and the outcome variable is grade ≥2 leukopenia and neutropenia, with volume units in 10 ml.

Table 3 Sensitivity analysis of ≥2 grade leukopenia and neutropenia related to each irradiated CV_X based on logistic regression

		≥2 grade leukopenia					≥2 grade neutropenia
	model	OR	95%CI		P-value		OR	95%CI	P-value
CV₅	Not adjusted	1.05		1.00，1.10		0.048	1.08	1.02,1.14	0.006
	model I	1.11		1.03，1.18		0.004	1.11	1.04,1.19	0.003
	model II	1.22		1.09，1.37		<0.001	1.13	1.04,1.24	0.006
CV₁₀	Not adjusted	1.04		0.99，1.09		0.106	1.07	1.01,1.13	0.015
	model I	1.10		1.02，1.18		0.009	1.10	1.03,1.19	0.006
	model II	1.24		1.10，1.40		<0.001	1.12	1.02,1.24	0.016
CV₁₅	Not adjusted	1.03		0.98，1.08		0.204	1.06	1.00,1.12	0.034
	model I	1.08		1.01，1.16		0.025	1.10	1.02,1.18	0.013
	model II	1.19		1.07，1.33		0.002	1.11	1.01,1.23	0.028
CV₂₀	Not adjusted	1.03		0.98，1.08		0.305	1.06	1.00,1.12	0.051
	model I	1.06		0.99，1.13		0.085	1.08	1.00,1.16	0.057
	model II	1.14		1.03，1.26		0.013	1.09	0.99,1.20	0.083
CV₂₅	Not adjusted	1.00		0.95，1.05		0.921	1.03	0.97,1.08	0.326
	model I	1.01		0.95，1.07		0.770	1.03	0.97,1.10	0.303
	model II	1.03		0.95，1.12		0.462	1.01	0.94,1.10	0.727
CV₃₀	Not adjusted	0.99		0.94，1.04		0.617	1.01	0.96.1.06	0.731
	model I	0.99		0.93，1.06		0.791	1.01	0.95,1.08	0.764
	model II	1.02		0.94，1.10		0.710	0.99	0.92,1.06	0.763
CV₃₅	Not adjusted	0.98		0.92，1.04		0.467	0.99	0.93,1.06	0.878
	model I	0.98		0.90，1.06		0.562	0.99	0.92,1.07	0.797
	model II	1.00		0.91，1.11		0.955	0.96	0.87,1.05	0.368
CV₄₀	Not adjusted	0.98		0.90，1.06		0.628	0.99	0.90,1.08	0.768
	model I	0.99		0.89，1.09		0.775	0.98	0.88,1.08	0.685
	model II	1.02		0.90，1.17		0.721	0.94	0.83,1.07	0.349

The exposure variable is the total pelvic bone marrow volume (CV₅-CV₄₀), and the outcome variable is grade ≥2 leukopenia and neutropenia, with volume units in 10 ml.

Table 4 Sensitivity analysis of ≥2 grade leukopenia and neutropenia related to each irradiated SV_X based on logistic regression

		≥2 grade leukopenia			≥2 grade neutropenia
	model	OR	95%CI	P-value	OR	95%CI	P-value
SV₅	Not adjusted	1.00	0.90，1.11	0.986	1.09	0.96,1.22	0.172
	model I	1.01	0.88，1.15	0.901	1.11	0.97,1.27	0.137
	model II	0.74	0.59，0.94	0.012	0.93	0.77,1.13	0.477
SV₁₀	Not adjusted	0.98	0.88，1.10	0.777	1.08	0.96,1.21	0.205
	model I	0.99	0.87，1.13	0.892	1.11	0.97,1.27	0.136
	model II	0.72	0.57，0.92	0.007	0.95	0.78,1.14	0.569
SV₁₅	Not adjusted	0.98	0.88，1.09	0.719	1.07	0.95,1.20	0.265
	model I	0.98	0.86，1.12	0.759	1.09	0.96,1.25	0.196
	model II	0.76	0.61，0.95	0.015	0.95	0.79,1.15	0.590
SV₂₀	Not adjusted	0.99	0.89，1.10	0.823	1.07	0.96,1.20	0.210
	model I	0.99	0.87，1.12	0.870	1.10	0.97,1.25	0.147
	model II	0.83	0.68，1.01	0.062	0.99	0.83,1.18	0.894
SV₂₅	Not adjusted	0.98	0.88，1.08	0.669	1.06	0.95,1.19	0.263
	model I	0.97	0.86，1.10	0.638	1.08	0.96,1.23	0.197
	model II	0.94	0.80，1.10	0.414	1.07	0.92,1.24	0.385
SV₃₀	Not adjusted	0.96	0.87，1.06	0.438	1.06	0.95,1.18	0.297
	model I	0.94	0.83，1.06	0.330	1.07	0.95,1.21	0.282
	model II	0.92	0.79，1.08	0.318	1.08	0.94,1.25	0.280
SV₃₅	Not adjusted	0.97	0.87，1.07	0.526	1.06	0.95,1.18	0.312
	model I	0.94	0.82，1.06	0.309	1.06	0.93,1.20	0.371
	model II	0.93	0.79，1.10	0.402	1.10	0.95,1.28	0.216
SV₄₀	Not adjusted	0.98	0.87，1.09	0.656	1.05	0.94,1.18	0.409
	model I	0.94	0.82，1.08	0.374	1.04	0.91,1.19	0.555
	model II	0.92	0.78，1.10	0.361	1.09	0.92,1.29	0.303

The exposure variable is the total pelvic bone marrow volume (SV₅-SV₄₀), and the outcome variable is grade ≥2 leukopenia and neutropenia, with volume units in 10 ml.

Table 5 Sensitivity analysis of ≥2 grade leukopenia and neutropenia related to each irradiated FV_X based on logistic regression

		≥2 grade leukopenia					≥2 grade neutropenia
	model	OR	95%CI			P-value	OR		95%CI		P-value
FV₅	Not adjusted	0.96		0.83, 1.11	0.252		0.52	0.13,2.78		0.574
	model I	0.99		0.85, 1.22	0.731		0.85	0.17,3.20		0.554
	model II	1.09		0.76, 1.52	0.302		1.42	0.10,6.25		0.920
FV₁₀	Not adjusted	0.94		0.73, 1.15	0.244		0.42	0.11,1.86		0.290
	model I	0.98		0.76, 1.15	0.762		0.64	0.11,6.10		0.702
	model II	1.11		0.84, 1.37	0.682		2.33	0.15,4.21		0.731
FV₁₅	Not adjusted	0.87		0.83, 1.08	0.122		1.02	0.72,1.44		0.334
	model I	0.88		0.63, 1.18	0.466		1.01	0.90,1.07		0.266
	model II	0.94		0.77, 1.41	0.607		1.07	0.98,1.03		0.134
FV₂₀	Not adjusted	0.91		0.74, 1.04	0.072		1.00	0.96,1.02		0.547
	model I	0.87		0.79, 1.08	0.152		1.00	0.97,1.06		0.843
	model II	0.81		0.81, 1.07	0.188		1.01	1.00,1.05		0.122
FV₂₅	Not adjusted	0.84		0.78, 1.01	0.058		0.97	0.93,1.01		0.227
	model I	0.81		0.49, 1.01	0.089		1.00	0.97,1.02		0.555
	model II	0.71		0.52, 1.00	0.052		1.01	0.98,1.09		0.821

The exposure variable is the total pelvic bone marrow volume (FV₅-FV₂₅), and the outcome variable is grade ≥2 leukopenia and neutropenia, with volume units in 10 ml.FV had almost no bone marrow volume receiving a dose exceeding 30 Gy.

3.Nonlinear fitting and threshold effect analysis: generalised additive models showed a nonlinear relationship between CV₅-CV₁₅ and ≥2 grade leukopenia, and a nonlinear relationship between CV₅-CV₁₀ and ≥2 grade neutropenia (Fig. 2).This study compared the logistic regression model (standard linear regression) with a 3-segment logistic regression model (piecewise linear regression).Using generalised additive model (GAM) and recursive algorithm, the inflection point values of CV₅, CV₁₀ and CV₁₅ for ≥2 grade leukopenia were obtained as (530 ml, 800 ml), (483 ml,745 ml) and (413 ml,696 ml) respectively, and for ≥2 grade neutropenia the CV₅ and CV₁₀ inflection point values were (528ml,800ml), (495ml,762ml) respectively.Where 483ml is the risk threshold for CV₁₀ corresponding to ≥2 grade leukopenia. When CV₁₀ was below the threshold, no significant correlation was observed with ≥2 grade leukopenia; when CV₁₀ exceeded the threshold, the risk of ≥2 grade leukopenia increased significantly with increasing CV, with an OR (95% CI) of 1.30 (1.11, 1.53) and a p-value of 0.001.The rest of the thresholds were all interpreted by similar reasoning (Table 6-7).

Table 6 Three piecewise linear regression of CV_X and ≥2 grade leukopenia

	CV₅			CV₁₀			CV₁₅
	OR	95%CI	P-value	OR	95%CI	P-value	OR	95%CI	P-value
standard linear regression	1.22	1.09,1.37	<0.001	1.24	1.10,1.40	<0.001	1.19	1.07,1.33	0.002
piecewise linear regression
inflection points（a，b）	53.0,80.0			48.3,74.5			41.3,69.6
V_x＜a	2.33	0.75,7.23	0.145	2.56	0.59,11.02	0.208	4.20	0.14,127.58	0.411
b ≥V_x≥ a	1.26	1.09,1.46	0.002	1.30	1.11,1.53	0.001	1.20	1.05,1.38	0.008
V_x＞b	2.40	0.61,9.36	0.209	1.68	0.89,3.18	0.110	2.15	0.79,5.84	0.135
P for Log-likelihood ratio test	0.016			0.027			0.044

The exposure variable is the coxal bone dose volume (CV_X), and the outcome variable is grade ≥2 leukopenia. The values “a” and “b” represent the smaller and larger inflection points, respectively. volume units in 10 ml.

Table 7 Three piecewise linear regression of CV_X and ≥2 grade neutropenia

	CV5			CV10
	OR	95%CI	P-value	OR	95%CI	P-value
standard linear regression	1.13	1.04,1.24	0.006	1.12	1.02,1.24	0.016
piecewise linear regression
inflection points（a，b）	52.8,80.0			49.5,76.2
V_x＜a	5.97	0.06,618.7	0.451	2.37	0.43,13.26	0.324
b ≥V_x≥ a	1.14	1.02,1.29	0.027	1.14	1.01,1.27	0.029
V_x＞b	1.60	0.68,3.74	0.282	1.95	0.39,9.78	0.416
P for Log-likelihood ratio test	0.050			0.033

The exposure variable is the coxal bone dose volume (CV_X), and the outcome variable is grade ≥2 neutropenia. The values “a” and “b” represent the smaller and larger inflection points, respectively. volume units in 10 ml.

The exposure variable is the Coxal bone dose volume (CV_X), volume units in 10 ml.

Figure 2 The smooth curvefits between the ≥2 grade leukopenia and neutropenia with coxal volume base on generalized additive model (GAM)

4.Analysis of CV_X versus SV_X at the same irradiated dose volume by linear regression resulted in a significant negative linear correlation at all dose volumes, as shown in Table 8.

Table 8 Linear regression analysis of CV and SV at the same dose

CV	β	95%CI	P值
CV₅	-0.52	-0.72, -0.25	<0.001
CV₁₀	-0.44	-0.62, -0.20	<0.001
CV₁₅	-0.37	-0.60, -0.09	0.032
CV₂₀	-0.31	-0.09, -0.37	<0.001
CV₂₅	-0.48	-0.76, -0.07	<0.001
CV₃₀	-1.10	-1.44, -0.82	<0.001
CV₃₅	-1.07	-1.17, -0.90	<0.001
CV₄₀	-1.01	-1.05, -0.96	<0.001

The adjusted variables included age, gender, BMI, T stage, N stage, chemotherapy regimen, and TVx, with volume units in 10 ml.

Acute hematologic toxicity (HT) is a common adverse reaction in patients undergoing concurrent chemoradiotherapy (CCRT) for locally advanced rectal cancer (LARC). Frequent or severe acute HT can increase the risk of infection and even delay treatment, ultimately affecting treatment efficacy (9, 10). In recent years, total neoadjuvant therapy (TNT) and neoadjuvant radiotherapy with immune checkpoint inhibitors have emerged as new trends in the preoperative treatment of locally advanced rectal cancer (17-20). However, in these treatment modalities, the increased intensity of chemotherapy administered concurrently or sequentially with preoperative radiotherapy also increases the risk of acute HT in patients (6, 8, 21). Therefore, minimizing the risk of hematologic adverse reactions during the course of chemoradiotherapy is beneficial for patients to successfully complete the preoperative neoadjuvant treatment. The aim of this study was to explore the real relationship between bone marrow dose volume during pelvic irradiation and hematologic adverse reactions, and to minimize the risk of pelvic radiotherapy-related hematologic adverse reactions by optimizing the radiotherapy plan.

In this study, univariate and multivariate logistic regression analysis (Model I) showed a significant correlation between acute HT and low dose volume of bone marrow (TV₅-TV₁₀, CV₅-CV₁₅). When adjusting for SV as a covariate in the multivariate logistic analysis, a varying degree of correlation was observed between low dose volumes of TV or CV and different types of acute HT. This finding is consistent with the results of Huang et al. (11)in rectal cancer and Mell et al. (14, 15) in cervical and anal cancer, which suggest that low dose volume of pelvic bone marrow is an important factor contributing to acute HT. However, the study by Yang et al. (9) found a correlation between high dose volume of sacrum and acute HT, which may be attributed to the use of different radiotherapy techniques in their study. Yang et al.(9) used 3D-CRT technique, which has a larger high dose volume compared to the IMRT technique used in our study (22). Additionally, there are differences in target distribution between these two radiotherapy techniques, suggesting that the factors influencing acute HT may differ between the two techniques. However, further research is needed to explore the specific reasons behind these differences. Compared to IMRT and 3D-CRT techniques, newer generation radiotherapy techniques (VMAT, TOMO) can effectively reduce high dose volume outside the target area while increasing the low dose volume of pelvic bone marrow (22, 23). Therefore, the aforementioned research results suggest that clinicians should pay more attention to controlling the low dose volume of pelvic bone marrow in patients undergoing pelvic radiotherapy using newer generation techniques.

Unlike previous studies, this study is the first to include the corresponding SV as a covariate in the multivariate logistic regression analysis of the relationship between bone marrow dose volume and acute HT. The reason for including SV as a covariate is that when we divided the total pelvic bone marrow into three parts (CV,SV,FV) and analyzed their respective correlations with acute HT, we found that CV was a risk factor for acute HT, with OR values generally >1 (as shown in Table 3), while the OR values for SV were all <1 (as shown in Table 4). This indicates that CV and SV have distinct and potentially interacting effects on acute HT. Through linear regression analysis of the relationship between CV and SV, the results showed a significant negative linear correlation between CV and SV (as shown in Table 7). By analyzing the dose and weight of different field angles, we were able to explain this result. Under the condition of a constant prescribed dose in the target area, the dose passing through the posterior field of the sacrum (mainly at 0 degrees) and the dose passing through the bilateral fields of the hip bone were negatively correlated. The hematopoietic function of the hip bone marrow is significantly stronger than that of the sacral bone marrow (24). Therefore, when the sacral bone marrow receives a higher dose (with a larger weight for the 0-degree field), the dose received by the hip bone marrow will correspondingly decrease, resulting in better overall protection of the pelvic bone marrow's hematopoietic function. Therefore, when analyzing the correlation between CV and acute HT, adjusting for the corresponding SV as a covariate can more accurately reflect the relationship between CV and acute HT.

Huang et al.(11) and Mell et al. (14, 15) analyzed the bone marrow dose volume parameters related to acute HT in IMRT, but they only analyzed the overall pelvic bone marrow. Similarly, Yang et al. (9) did not perform a multivariate analysis of the effects of different pelvic bone marrow regions on acute HT. In this study, we conducted a multivariate logistic regression analysis of the three components of TV (hip bone, sacral bone, femoral bone) and acute HT. The results showed a significant correlation between low dose levels of CV (CV₅, CV₁₀, CV₁₅, CV₂₀) and the risk of acute HT, indicating that CV is a risk factor for acute HT. On the other hand, low dose levels of SV (SV₅, SV₁₀, SV₁₅) were negatively correlated with acute HT, and FV showed no significant correlation with acute HT. This suggests that CV is the determining factor for acute HT in the pelvic bone marrow. Based on these results, CV is a better indicator for predicting the risk of acute HT compared to TV.

To enhance the clinical value of this study, further analysis was conducted using the generalized additive model and segmented regression to explore the potential nonlinear relationship and threshold effect between CV and acute HT. The generalized additive model showed a nonlinear relationship between CV₅, CV₁₀, CV₁₅, and grade ≥2 leukopenia. CV₅, CV₁₀also exhibited a nonlinear relationship with grade ≥2 neutropenia. In the segmented logistic regression analysis, the log odds ratio (OR) was significant (P<0.05), confirming the presence of threshold effects in these nonlinear relationships. The analysis results showed that there was no significant correlation between CV₁₀ and grade ≥2 leukopenia when CV₁₀was below the threshold of 483 ml. However, once CV₁₀ exceeded the threshold, an increase in CV significantly increased the risk of grade ≥2 leukopenia, with an OR (95% CI) of 1.30 (1.11, 1.53). Additionally, the CV thresholds for the two types of acute HT were very close for the same dose. Based on these findings, in optimizing the intensity-modulated radiation therapy plan for rectal cancer, it is reasonable to consider using a CV₁₀ threshold of 483 ml as a reference value to effectively reduce the risk of grade ≥2 acute HT while maximizing the utilization of normal tissue tolerance. However, it should be noted that the sample size of this study was small, and regional and ethnic differences may have an impact on the thresholds.

In neoadjuvant IMRT for rectal cancer, CV is a better indicator for predicting acute HT compared to TV. CV, particularly at low dose levels (CV₅, CV₁₀, CV₁₅, CV₂₀), is closely associated with acute HT. The threshold value identified in this study (CV10=483 ml) can serve as a reference for optimizing radiation therapy plans to reduce the risk of acute HT in pelvic radiation for rectal cancer.

3D-CRT

3-dimensional conformal radiation therapy

5-Fu

5-fluorouracil

ANC

absolute neutrophil count

BMI

Body mass index

CBM

Coxal bone marrow

Confidence Interval

CCRT

concurrent chemoradiotherapy

DVH

dose volume histogram

FBM

Femoral bone marrow

GAM

generalised additive model

hemoglobin

hematological toxicity

IMRT

intensity-modulated radiotherapy

KPS

Karnofsky

LARC

locally advanced rectal cancer

NCI-CTC

National Cancer Institute Common Toxicity Criteria

OARs

organs at risk

odds ratio

PBM

Pelvic bone marrow

pCR

pathologic complete response

PLT

platelet

RTOG

Radiation Therapy Oncology Group

SBM

Sacrum bone marrow

TNT

Total neoadjuvant therapy

Total volume

VMAT

volumetric modulated arc therapy

WBC

white blood cells

Acknowledgments

Many thanks to staff members Haojia Zhang and Haijie Jin in Guizhou Cancer Hospital Cancer Therapy Center for they support and assistance with my research project, as well as for guidance on target area outlining in this study.

Authors' Contribution：

Sicheng Huang and Gang Wang proposed the research idea, designed the study protocol, implemented the study and wrote the paper;Xiaokai Li, Yun Li,Guangyi Jiang and Songyou Wu were responsible for collecting clinical data and outlining the target area; Sicheng Huang, Gang Wang and Wenling Wang participated in the discussion and revision of the paper.

Funding

This study was supported by the Joint Fund (QianKeHeJiChu-ZK[2024]YiBan 360) of Guizhou Science and Technology Department. The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Availability of data and materials

The datasets generated and/or analyzed during the current study are not publicly available since the participants did not consent to share the data with third parties.

Ethics approval and consent to participate

The institutional ethics committee granted ethics approval for this study (SL-201507004; Guizhou Cancer Hospital, Guiyang, People’s Republic of China).

Consent for publication

All authors consent to publication.

Competing interests

The authors declare that they have no competing interests.

Prolongation of the. disease-free interval in surgically treated rectal carcinoma. N Engl J Med. 1985;312(23):1465–72.
Krook JE, Moertel CG, Gunderson LL, Wieand HS, Collins RT, Beart RW, et al. Effective surgical adjuvant therapy for high-risk rectal carcinoma. N Engl J Med. 1991;324(11):709–15.
Sauer R, Becker H, Hohenberger W, Rödel C, Wittekind C, Fietkau R, et al. Preoperative versus postoperative chemoradiotherapy for rectal cancer. N Engl J Med. 2004;351(17):1731–40.
Sauer R, Liersch T, Merkel S, Fietkau R, Hohenberger W, Hess C, et al. Preoperative versus postoperative chemoradiotherapy for locally advanced rectal cancer: results of the German CAO/ARO/AIO-94 randomized phase III trial after a median follow-up of 11 years. J Clin Oncol. 2012;30(16):1926–33.
Gérard JP, Conroy T, Bonnetain F, Bouché O, Chapet O, Closon-Dejardin MT, et al. Preoperative radiotherapy with or without concurrent fluorouracil and leucovorin in T3-4 rectal cancers: results of FFCD 9203. J Clin Oncol. 2006;24(28):4620–5.
Deng Y, Chi P, Lan P, Wang L, Chen W, Cui L, et al. Modified FOLFOX6 With or Without Radiation Versus Fluorouracil and Leucovorin With Radiation in Neoadjuvant Treatment of Locally Advanced Rectal Cancer: Initial Results of the Chinese FOWARC Multicenter, Open-Label, Randomized Three-Arm Phase III Trial. J Clin Oncol. 2016;34(27):3300–7.
Wan J, Liu K, Li K, Li G, Zhang Z. Can dosimetric parameters predict acute hematologic toxicity in rectal cancer patients treated with intensity-modulated pelvic radiotherapy? Radiat Oncol. 2015;10:162.
Zhu J, Liu A, Sun X, Liu L, Zhu Y, Zhang T, et al. Multicenter, Randomized, Phase III Trial of Neoadjuvant Chemoradiation With Capecitabine and Irinotecan Guided by UGT1A1 Status in Patients With Locally Advanced Rectal Cancer. J Clin Oncol. 2020;38(36):4231–9.
Yang TJ, Oh JH, Apte A, Son CH, Deasy JO, Goodman KA. Clinical and dosimetric predictors of acute hematologic toxicity in rectal cancer patients undergoing chemoradiotherapy. Radiother Oncol. 2014;113(1):29–34.
Jianyang W, Yuan T, Yuan T, Xin W, Ning L, Hua R, et al. A prospective phase II study of magnetic resonance imaging guided hematopoietical bone marrow-sparing intensity-modulated radiotherapy with concurrent chemotherapy for rectal cancer. Radiol Med. 2016;121(4):308–14.
Huang W, Dang J, Li Y, Cui HX, Lu WL, Jiang QF. Effect of Pelvic Bone Marrow Sparing Intensity Modulated Radiation Therapy on Acute Hematologic Toxicity in Rectal Cancer Patients Undergoing Chemo-Radiotherapy. Front Oncol. 2021;11:646211.
Albuquerque K, Giangreco D, Morrison C, Siddiqui M, Sinacore J, Potkul R, et al. Radiation-related predictors of hematologic toxicity after concurrent chemoradiation for cervical cancer and implications for bone marrow-sparing pelvic IMRT. Int J Radiat Oncol Biol Phys. 2011;79(4):1043–7.
Cheng JC, Bazan JG, Wu JK, Koong AC, Chang DT. Lumbosacral spine and marrow cavity modeling of acute hematologic toxicity in patients treated with intensity modulated radiation therapy for squamous cell carcinoma of the anal canal. Pract Radiat Oncol. 2014;4(3):198–206.
Mell LK, Kochanski JD, Roeske JC, Haslam JJ, Mehta N, Yamada SD, et al. Dosimetric predictors of acute hematologic toxicity in cervical cancer patients treated with concurrent cisplatin and intensity-modulated pelvic radiotherapy. Int J Radiat Oncol Biol Phys. 2006;66(5):1356–65.
Mell LK, Schomas DA, Salama JK, Devisetty K, Aydogan B, Miller RC, et al. Association between bone marrow dosimetric parameters and acute hematologic toxicity in anal cancer patients treated with concurrent chemotherapy and intensity-modulated radiotherapy. Int J Radiat Oncol Biol Phys. 2008;70(5):1431–7.
Kuntz L, Noel G. [Pelvic irradiation and hematopoietic toxicity: A review of the literature]. Cancer Radiother. 2021;25(1):77–91.
Cercek A, Roxburgh CSD, Strombom P, Smith JJ, Temple LKF, Nash GM, et al. Adoption of Total Neoadjuvant Therapy for Locally Advanced Rectal Cancer. JAMA Oncol. 2018;4(6):e180071.
Giunta EF, Bregni G, Pretta A, Deleporte A, Liberale G, Bali AM, et al. Total neoadjuvant therapy for rectal cancer: Making sense of the results from the RAPIDO and PRODIGE 23 trials. Cancer Treat Rev. 2021;96:102177.
Yang Z, Gao J, Zheng J, Han J, Li A, Liu G, et al. Efficacy and safety of PD-1 blockade plus long-course chemoradiotherapy in locally advanced rectal cancer (NECTAR): a multi-center phase 2 study. Signal Transduct Target Ther. 2024;9(1):56.
Rahma OE, Yothers G, Hong TS, Russell MM, You YN, Parker W, et al. Use of Total Neoadjuvant Therapy for Locally Advanced Rectal Cancer: Initial Results From the Pembrolizumab Arm of a Phase 2 Randomized Clinical Trial. JAMA Oncol. 2021;7(8):1225–30.
Wang X, Yu Y, Meng W, Jiang D, Deng X, Wu B, et al. Total neoadjuvant treatment (CAPOX plus radiotherapy) for patients with locally advanced rectal cancer with high risk factors: A phase 2 trial. Radiother Oncol. 2018;129(2):300–5.
Chang Y, Yang ZY, Li GL, Li Q, Yang Q, Fan JQ, et al. Correlations Between Radiation Dose in Bone Marrow and Hematological Toxicity in Patients With Cervical Cancer: A Comparison of 3DCRT, IMRT, and RapidARC. Int J Gynecol Cancer. 2016;26(4):770–6.
Kumar T, Schernberg A, Busato F, Laurans M, Fumagalli I, Dumas I, et al. Correlation between pelvic bone marrow radiation dose and acute hematological toxicity in cervical cancer patients treated with concurrent chemoradiation. Cancer Manag Res. 2019;11:6285–97.
Russell WJ, Yoshinagah, Antoku S, Mizuno M. Active bone marrow distribution in the adult. Br J Radiol. 1966;39(466):735–9.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Correlation between pelvic bone marrow dose volume parameters and acute hematologic toxicity in neoadjuvant intensity-modulated radiotherapy for locally advanced rectal cancer

Status:

Version 1

Abstract

Objective

Methods

Results

Conclusion

Figures

Background

Materials and methods

1.Clinical data:

2.Treatment plan:

3.Bone marrow region outlining and dose-volume parameters:

Results

Discussion

Conclusion

Abbreviations

Declarations

References

Additional Declarations

Status:

Version 1