Radiation-Related Predictors of Hematologic Toxicity After Concurrent Chemoradiation for Cervical Cancer and Implications for Bone Marrow–Sparing Pelvic IMRT

Purpose

To determine factors predictive for hematologic toxicity (HT) associated with concurrent chemoradiation for Stage II through IV cervical cancer.

Methods and Materials

The medical records of 40 women receiving concurrent chemoradiation for cervical cancer were reviewed. Hematologic toxicity was defined by use of Common Terminology Criteria for Adverse Events (version 3.0). Variables predicting for HT including age, body mass index, transfusions, and bone marrow volumes irradiated were included in the data analysis.

Results

Of the patients, 13 (32.5%) had Grade 0 or 1 HT and 27 (67.5%) had Grade 2 through 4 HT (HT2+). Multiple logistic regression analysis of potential predictors showed that only the volume of bone receiving 20 Gy (V20) for whole pelvic bone tended toward significance for predicting HT2+. A strong correlation was noted between HT2+ and V20 (r = 0.8, p < 0.0001). A partitioning analysis to predict HT2+ showed a cutoff value of 79.42% (approximately 80%) for V20 of whole pelvic bone. That is, if the V20 of the whole pelvis exceeds 80%, the risk of HT2+ developing increases by a factor (odds ratio) of 4.5 (95%, confidence interval, 1.08–18.69) (p < 0.05).

Conclusions

We have shown a correlation between bone marrow volume radiated and development of HT. This has implications for use of pelvic intensity-modulated radiation therapy, which can potentially decrease the volume of bone marrow radiated.

Introduction

The current standard of care for the management of advanced cervical carcinoma is concurrent chemoradiation therapy (CCRT) followed by brachytherapy 1, 2. The goal of these combined therapies is to maximize tumor cell death with the radiosensitizing effects of chemotherapy while minimizing tumor repopulation by completing therapy within 8 weeks. Although this treatment is effective and has been validated by Phase III randomized studies 3, 4, 5, 6, the combination of chemotherapy and radiation increases the risk of hematologic toxicity (HT), as shown in reviews of the subject 7, 8. This acute toxicity is more severe with combined therapy compared with radiation therapy (RT) alone (8).

Patients who receive CCRT for cervical carcinoma have an increased risk of Grade 2 through 4 neutropenia, thrombocytopenia, and anemia developing. These women also have an increased risk of Grade 2 through 4 acute gastrointestinal and genitourinary toxicities along with the prolongation of median treatment time (7). Patterns-of-care studies have shown that treatment prolongation of 1 day corresponds roughly to a 1% decrease in local control (9). The presence of significant neutropenia or thrombocytopenia (Grade 3 or 4) can result in mandatory prolongation of radiation treatment, which has been shown to be detrimental for local tumor control.

Hematologic toxicity has been particularly noted in women undergoing pelvic RT for cervical cancer, compared with other locations, because of the increased radiosensitivity of pelvic bone marrow (BM), which is in the radiation field (10). Moreover, approximately 40% of the total-body BM reserve lies within the pelvic bones (11). This radiation–related risk for HT is in addition to myelotoxic effects of chemotherapy agents used in these regimens. Conventional two-dimensional four-field pelvic radiotherapy as was used in the previously mentioned Phase III studies irradiates large volumes of BM, which can be a contributory factor for HT. Using computed tomography (CT) image–based planning, one can deliver three-dimensional conformal radiation (3D-CRT) to the clinical target volume (CTV), which may reduce the volume of BM in the field. A series of publications from a single institution have studied the impact of whole-pelvis intensity-modulated radiation therapy (IMRT) and concurrent chemotherapy and showed a relationship between HT and volume of BM radiated to low doses 12, 13, 14.

Because IMRT is known to result in large regions of low-dose radiation and BM stem cells are exquisitely radiosensitive, it is possible that the relationships derived by these studies may not apply to whole-pelvis 3D-CRT, where the gradient between moderate and low isodose levels is quite sharp. Hence we chose to evaluate a series of patients who have received whole-pelvis 3D-CRT with concurrent chemotherapy and correlate several parameters to determine which of these are important predictors of HT. Our goal was to define these important parameters and their relationship to HT. If these factors are modifiable, one could potentially reduce HT, thereby avoiding prolongation of total treatment completion time.