Dose and polymorphic genes xrcc1, xrcc3, gst play a role in the risk of developing erythema in breast cancer patients following single shot partial breast irradiation after conservative surgery

From March 2006 to January 2008, patients who underwent BCS and sentinel node biopsy and/or axillary dissection for early breast adenocarcinoma and who met the eligibility criteria were treated in prone position. An adjuvant single dose 3D conformal radiation therapy (3D-CRT) APBI schedule was administered to the Index Area. The eligibility criteria included: Age ≥ 48 years with a life expectancy of at least 5 years, post-menopausal status, histologically proven, non lobular, adenocarcinoma of the breast, primary tumours ≤ 3 cm, negative surgical margins (≥ 2 mm), negative sentinel nodes or < 4 positive axillary nodes, no extra-capsular extension, no previous radiotherapy. The exclusion criteria comprised of the following: multicentric disease, extended intraductal component (EIC > 25%), Paget's disease, lobular adenocarcinoma, distant metastases. All the above criteria enable the identification of patients at low risk of local recurrence. The study was conducted in accordance with the Helsinki Declaration. Each patient was informed in advance about the study protocol in written form (informed consent) and verbal. The patient was given ample opportunity to request relevant information regarding the study and decide with full autonomy whether to participate in the protocol. The protocol was approved by the local Ethics and Scientific Committee of the Regina Elena Italian National Cancer Institute (reference number IFO-84/10).

From March 2006 to January 2008, 57 patients matched the eligibility criteria and provided a written informed consent. Instead of given an equivalent conventional treatment consisting of 50 Gy (25 daily fractions of 2 Gy in 5-6 week treatments) a single dose was used. The dose of 18 Gy was used in the first 4 patients to test the feasibility of the protocol and whether it was well tolerateda dose a of 21 Gy was administered to the other patients. The dose was delivered to a portion of breast parenchyma corresponding to the tumour bed after surgery. Dose volume constraints were used in order to reduce the dose to normal breast tissue and skin. The major technical details of our approach were reported previously in a separate paper [23].

The SSPBI trial was designed as a prospective Phase II single-arm study. The use of a single dose tumour bed is expected to be very effective in terms of tumour control, but it could increase the incidence of radiation induced erythema. Therefore, we assumed that a decreased DNA repair capability, as well as a reduced detoxification of the damage caused by oxidative stress could explain the increased acute toxicity, i.e. a higher incidence of erythema after a single dose. It is for this reason we decided to investigate SNPs of genes involved in antioxidant and DNA damage repair pathways such as GST, XRCC1, XRCC3 and RAD51.

We assumed an erythema rate of 20% and 54% in patient groups at low and high risk, respectively, (groups were identified based on the absence/presence of the above polymorphisms alone or in combination). Thus the minimum sample size was 56 patients with α = 0.05, 2-tailed test and a power of the study of 80%.

Acute toxicity (i.e. erythema), was the end-point described in the analysis and was assessed using the Regina Elena Italian National Cancer Institute common terminology criteria for adverse events (CTCAE, version 3.0) [24], and was defined as acute if it occurred within the first six months after radiotherapy. Grade refers to the severity of the Adverse Event (AE). Briefly, Grade 1 indicates faint erythema or dry desquamation, G2 moderate to brisk erythema, G3 moist desquamation other than skin folds and creases and G4 skin necrosis of ulceration of full thickness dermis.

All subjects enrolled in the study provided samples of blood, approximately 5 ml, in sterile tubes containing ethylenediaminetetracetic acid (EDTA). Whole blood samples for DNA analyses were immediately frozen at -80°C until processing.

Genomic DNA was isolated from cells in the venous blood using QIAmp kit (QIAmp DNA blood Mini Kit, Qiagen, Valencia, CA), following the manufacturer's instructions, and the DNA quality was evaluated by the spectrophotometer analysis (NanoDrop instrument). Specific primers included in the Radiotherapy Response kit (Diathec Company, Italy) were used in the DNA amplification experiments. The desired DNA was amplified by Real-Time PCR using Rotorgene Instrument (Corbett) following polymerase chain reaction (PCR) conditions provided by the manufacturer. The following polymorphisms were evaluated: XRCC3 C18067T (Thr241Met), XRCC3 A4541G (5'-UTR untranslated region), XRCC1 G28152A (Arg399Gln), GSTP1 A313G (Ile105Val) RAD51 G135C (untranslated region).

Other polymorphisms such as GSTA1 and XRCC1 Arg194Trp, which were not included in the above kit, were carried out and tested. The ad-hoc designed primers for these genes are illustrated in Table 1 and PCR reactions were carried out in a volume of 50 μl containing: 10 mM Takara deoxynucleotide triphosphate (dNTP) mixture, 20 pmol primers, approximately 30 ng of DNA template, Takara 5XR-PCR Buffer (Mg²⁺free), Takara 50 mM Mg²⁺, EvaGreen TM Dye 20 × and Takara Ex TaqR-PCR Custom (5U/ul). Reaction conditions were as follows: initial denaturation at 95°C for 3 min then 35-cycles of denaturation at 95°C for 30 sec; annealing at 56°C (GSTA1) or at 62°C (XRCC1 Arg194Trp) for 30 sec; elongation at 72°C for 30 sec and final extension at 60°C for 5 min, then 5-cycles of green channel signal acquisition at 60°C for 30 sec. PCR products were evaluated on 2.5 % agarose gel stained with ethidium bromide. In addition, one control for each genotype of GSTA1 and XRCC1 Arg194Trp was generated.

The target sequence containing the polymorphic site was amplified using standard PCR conditions in which one of the primers was biotinylated. A single-strand template was generated by removing the non-biotinylated strand on streptavidin coated beads (matrix). The bound DNA template on the affinity matrix was separated by denaturation in NaOH and used for the following synthesis of a short strand (10-15 bases) of DNA adjacent to the SNP site. The specific sequence primers used for GSTA1 and XRCC1 Arg194Trp were also reported in Table 1. The polymorphisms were analyzed using Pyrosequencing technologies (instrument PyroMark MD-Biotage, Uppsala, Sweden) according to a previously published method [25, 26].

In order to determine predictors of acute toxicity, retrospectively, we evaluated the following polymorphisms: XRCC3 C18067T (Thr241Met), XRCC3 A4541G (5'-UTR), XRCC1 G28152A (Arg399Gln), GSTP1 A313G (Ile105Val) RAD51, G135C (untranslated region) GSTA1 and XRCC1 Arg194Trp. In addition, we also analyzed combined genotypes according to the literature.

Odds ratios (ORs) and 95% confidence intervals (CIs), Chi-squared and Fisher exact (2-sided) tests were calculated. A OR > 1.0 indicates an increased risk of erythema in patients with a polymorphic gene.

Patient characteristics, polymorphisms and dosimetric data, selected as significant by the Univariate analysis, were included in a logistic regression analysis in the multivariate model, which included mean dose to whole breast (WB) and to skin, age at the time of treatment and all polymorphisms.

Statistically significant associations between polymorphisms and side effects were indicated by a p-value < 0.05. The sample size was calculated with Sample Power, the odds risks with an R-package, and the Forest plots with Medcalc.

Pyrosequencing is a sequencing method performed by synthesis, a simple to use technique for accurate and consistent analysis of large numbers of short to medium length DNA sequences. Each result, called a "pyrogram", indicates a determined polymorphic position, providing useful information to identify a category of patients with increased risk of toxicity.

For GSTA1 and XRCC1 Arg194Trp polymorphisms that are not included in the Radiotherapy kit, a method assay to validate them was developed and the typical pyrograms obtained are depicted in Figure 1. In this figure, the peak is proportional to the amount of nucleotides incorporated. The sequencing events are obtained in real time and the resulting genotypes of GSTA1 and XRCC1 Arg194Trp are depicted inside the yellow area. Each gene can generate three distinct pyrosequencing patterns, corresponding to homozygote wild type, heterozygote or homozygote mutant. The polymorphism was determined and the results are shown in Table 3.

Acute erythema was observed in 19 patients (33%), divided in Grade 1 (G1) in 11 (19%) and Grade 2 (G2) in 8 (14%) patients. No erythema was detected in 38 patients (67%). A statistically significant correlation was found between the ≥G1 erythema and the mean dose to WB (p = 0.008).

The ORs of any grade of erythema (with 95% CI and p-values) are reported in Table 4, by distinguishing the patients on the basis of the polymorphic genes. We found a significant incidence of erythema rate in patients with the polymorphic variant (AA i.e. wt) of XRCC1 Arg194Trp with greater odds (OR = 8.07; 95% CI, 1.02-373.8). The allelic variant wt of XRCC3 Thr241Met is associated with a lower rate of erythema (OR = 0.29; 95% CI, 0.05-1.29). A toxic role of allelic variant (wt/het i.e AA/Aa) of XRCC3A4541G-5'UTR was also found (OR = 0.49; 95% CI, 0.01-40.29).

Our results show a level that is not statistically significant in the protection towards developing an erythema, i.e. lower odds (OR = 0.3; 95% CI, 0.08-1.08), in the presence of at least one allelic variant (mut/het i.e. aa/Aa) of XRCC1 Arg194Trp or of GSTP1 (Ile105Val). The presence of the polymorphic variant mut/het of XRCC1 Arg194Trp and wt of XRCC3 Thr241Met results significantly in the protection towards developing an erythema with lower odds (OR = 0.2; 95% CI: 0.04-0.78).

Furthermore, the association between the polymorphic variant mut/het of XRCC1 Arg194Trp and the wt of XRCC1 Arg399Gln shows a protective role towards erythema (OR = 0.39; 95% CI: 0.1-1.35). While, the association between the polymorphic variant mut/het of XRCC1 Arg194Trp or wt of GSTA1 results in a higher risk of erythema (OR = 6.01; 95% CI, 1.16-61.04).

Forest plots (Figure 2 panels A-E) show the acute skin toxicity observed in our cohort and in literature, by distinguishing the patients based on the presence of alleles of some polymorphic genes. A protective role (as trend) of mut/het XRCC1 Arg194Trp was found, in contrast to data by Mangoni et al. [12] that reported a slightly toxic effect (Figure 2A). A significant function of wt XRCC1 Arg399Gln as toxic agent was observed according to data reported by Popanda et al. [13] and Mangoni et al. [12] (Figure 2B). Next, the trend as a toxic factor of the wt XRCC3 Thr241Met was found, according to Chang-Claude et al. [14]; as well as of wt GSTA1, as reported by Ambrosone et al. [15] who described this gene as protective agent (Figure 2C-D). Further, we found a protective role as trend for wt GSTP1 in contrast to Ambrosone et al. [15] (Figure 2E). In fact, no statistical significance for wt GSTP1 was found to distinguish acute toxicity in patients in either of the studies. Finally, no correlation was found between acute toxicity and mut/het RAD51. Multivariate analysis confirmed the mean dose to WB (p = 0.002), the presence of mut/het XRCC1 Arg194Trp or wt XRCC3 Thr241Met (p = 0.006) and the presence of mut/het XRCC1 Arg194Trp or wt GSTA1 (p = 0.031) as predictors of erythema. In addition, when identifying the low and high risk groups, according to the absence or presence of mut/het XRCC1 Arg194Trp or wt XRCC3 Thr241Met, the observed erythema in our cohort was 20% and 59%, respectively. Assuming an α = 0.05 and a 2-tailed test, the power of the study was 78% with 38 and 19 patients in low and high risk groups respectively; when assuming an α = 0.05 and a 1-tailed test was used the power of the study was 80%.