Radiation Complications and Their Management

Introduction

This chapter discusses the potential complications associated with the use of radiotherapy for breast cancer. These may occur in the weeks during or after radiation, decades after treatment, or even years after treatment in some cases. The time course for the different complications discussed is different from each other—for example, skin complications typically occur during or shortly after radiation, while cardiac effects are expected years after treatment. The time that a particular complication may typically appear could also vary even within the same organ—for example, there are differences between subacute and chronic complications for lung tissues. Complications may also be more or less common depending on the specific radiation details—for example, cardiac risks may be minimal for a right-sided cancer compared to a left-sided cancer. Improvements in radiation therapy in the past decade have come in areas of physical equipment, 3D computer simulation and automated design, and understanding of radiation fractionation. All of these radiation advancements fortunately have led to a decrease in the incidence of many of these complications in patients treated today compared with those treated in the past with outdated techniques and equipment.

Such possible complications, from most to least common, include fatigue; myelosuppression; radiation dermatitis; alterations in the cosmetic appearance of the breast and local soft tissue symptoms; chest wall or soft tissue complications, including rib fractures and brachial plexopathy; pulmonary effects; cardiac complications; and radiation-related second malignancies ( Table 49.1 ). (The effects of irradiation on the risks of arm edema and complications after breast reconstructive surgery are discussed in other chapters.) The risk of complications in patients with collagen vascular disease is also discussed in this chapter.

Fatigue and Myelosuppression

Fatigue is a common side effect in women treated for breast cancer in general and also specifically in association with radiation. Baseline assessments of fatigue in breast cancer patients even before radiation show that approximately 30% or more report fatigue. Baseline fatigue has been associated with higher cancer stage and many commonly used medications, including pain or psychotropic medications. Fatigue generally increases weekly during radiation to plateau at weeks 4 to 6 but is usually mild and resolves or returns to pretreatment baseline within a few months of completing treatment. Radiation-related fatigue may be associated with decreased quality of life. However, in one randomized trial of breast-conserving surgery (BCS) with or without radiation, radiation was associated with increased levels of fatigue but not with global differences in quality of life. Fatigue remains common in 40% to 50% of women by the end of radiation, and can remain present even 1 year after completion of treatment. And in another long-term longitudinal study of breast cancer survivors, persistent fatigue was reported in approximately 20% of women up to 5 to 10 years from treatment. Fatigue is more common after chemotherapy and radiation compared with either treatment alone.

In managing the patient with fatigue, it is important to rule out other potential causes such as anemia, cardiovascular disease, depression, or hypothyroidism. There is no specific medical treatment for radiation-related fatigue. In a randomized double-blind clinical trial of vitamins or placebo during radiation for breast cancer, there was no improvement in radiation-related fatigue. Physical activity including aerobic and resistance exercise has been associated with reductions in patient-reported fatigue after radiation in randomized trials. In another randomized trial, patients assigned to up to three 60-minute classes of yoga or stretching exercises per week during their 6 weeks of radiation had improved fatigue at the end of treatment and 1 to 3 months after treatment compared with patients assigned to usual care (and offered these interventions after radiation). Activity in the structure of a regular community-based exercise program for 6 months beginning during or within 3 months of treatment was shown to reduce even long-term fatigue in participants. However, two comprehensive reviews and meta-analyses of randomized and nonrandomized controlled trials have not shown statistically significant improvements in fatigue or health-related quality of life with physical exercise compared with controls. Acupuncture has been associated with improvements in cancer-related fatigue in patients with breast cancer in one prospective randomized trial.

Myelosuppression is also very common during and for a few months after radiation, with the greatest effect on circulating lymphocytes and the least effect on platelets and hemoglobin levels. Breast irradiation directly treats a very small volume of total body bone marrow of the chest wall. Because myelosuppression is rarely of clinical significance, blood counts do not need to be routinely checked. Hematologic toxicities appear to be slightly greater in patients treated with chemotherapy after radiation therapy, compared with when chemotherapy is given before radiation which is more typical in modern clinical practice. In randomized trials of concurrent versus sequential chemotherapy and radiation, anemia or febrile neutropenia were more frequent with concurrent therapy. For the rare patient receiving concurrent chemotherapy, special attention should be given to signs of infection (local or systemic) or leukopenia. Although used less commonly in modern practice, cyclophosphamide-methotrexate-5-fluorouracil (CMF) chemotherapy is associated with a low risk for these complications when given concurrently with radiation. There is limited experience with radiation and other concurrent systemic chemotherapies, but in one trial the incidence of grade 3 to 4 neutropenia was much higher with concurrent radiation and paclitaxel but not felt to be different than expected from paclitaxel alone. Concurrent trastuzumab and breast radiation has not been associated with increased hematologic toxicity.

Radiation Dermatitis and Infections

Acute radiation dermatitis includes a clinical spectrum of signs and symptoms that usually develop slowly during the course of radiation but then gradually escalates as treatment progresses each week. These usually become most severe in the last week of radiation and then continue for 1 to 2 weeks after completion. Radiation dermatitis begins with generalized dryness and itching of the skin. This will commonly progress to erythema or hyperpigmentation, which may be asymptomatic or mildly painful. This erythema can be confused with acute inflammation of the breast that can also occur in the first weeks of treatment. Often a pruritic papular rash (folliculitis) may occur in the upper inner portion of the chest wall where there is overlap with previously sun-exposed regions of the neck and chest. Dry desquamation may progress to a moist, weeping reaction due to the breakdown of the skin integrity; the latter is particularly likely to be found in the skin folds of the underside of the breast and the axillary tail. Patients with such moist desquamation may also have substantial breast discomfort or pain and swelling. Moist desquamation is associated with a reduced global quality of life. Rarely, the most severe cases may progress to full-thickness skin ulceration and bleeding. The time course of recovery is generally weeks for immediate erythema or desquamation, whereas months may be required for the resolution of hyperpigmentation and swelling.

Acute dermatitis is usually scored by the National Cancer Institute’s Common Terminology Criteria for Adverse Events (CTCAE). Grade 1 dermatitis is faint erythema or dry desquamation; grade 2 is moderate to brisk erythema; patchy moist desquamation, mostly confined to skin folds and creases; moderate edema; grade 3 is moist desquamation other than skin folds and creases; bleeding induced by minor trauma or abrasion; grade 4 is skin necrosis or ulceration of full-thickness dermis; spontaneous bleeding from the involved site. The incidence of acute radiation dermatitis with BCS and modern radiation is approximately 3% grade 0, 35% grade 1, 60% grade 2, and 2% grade 3. The definitions used in CTCAE for dermatitis include adjectives such as mild, moderate, or mostly so are subjective and may allow substantial interobserver variation. Furthermore, each grade encompasses substantial variability in the severity of symptoms experienced by patients. For example, a small 1-cm nonpainful area of moist desquamation in the inframammary fold is scored as grade 2, as would be a 10-cm painful area requiring wound care or a treatment break. A Michigan skin dermatitis scale was created for breast radiation that includes a photonumeric scale for degrees of skin color changes and desquamation, and was highly consistent between raters and patient-reported symptoms. Fig. 49.1 shows examples of acute radiation skin toxicity.

Radiation dermatitis is most directly related to increased dose inhomogeneity or hot spots within the radiation treatment plan, which are themselves a function of increasing breast size or chest wall diameter. For this reason, moist desquamation is more common in women with large breasts than those with small breasts. Intensity-modulated radiation therapy (IMRT) has been associated with a decrease in the rates and severity of acute dermatitis, compared with conventional two-dimensionally compensated “wedged” tangential-field irradiation. For example, in one randomized trial, the dose homogeneity of the radiation plans was significantly improved (lower hot spots) for the patients receiving IMRT, and this correlated with less likelihood to develop moist desquamation. Improved methods of 3D conformal radiation have become standard practice because they are able to improve dose homogeneity using forward-based planning and “field-in-field” techniques that are less distinguishable from IMRT at less cost. Prone positioning may reduce toxicity previously seen in larger-breasted women by reducing their inframammary skin folds or decreasing chest wall separation. Whole breast hypofractionation, or using a larger daily dose but reduced total dose and number of fractions, that is now standard for most lumpectomy patients and has been associated with decreased acute radiation skin toxicity compared with conventional fractionation.

Randomized controlled trials of topical skin care products used with the goal of mitigating radiation dermatitis in breast cancer patients compared with an aqueous-type cream placebo have been mostly negative or of minimal clinical significance. These include studies of Biafine (Ortho-McNeil Pharmaceuticals, Raritan, NJ), hyaluronic acid, steroids, oil-based emulsions, calendula, glycerin, or miscellaneous natural ingredients. For example, in one randomized trial in postmastectomy radiation, where skin toxicity was higher than in lumpectomy patients due to the increased skin dose delivered, mometasone furoate 0.1% reduced moist desquamation; however, it remained very common (44% vs. 67%; P = 0.012), and there was no difference in patient-reported skin symptom outcomes. As shown in that trial, topical steroids have the most evidence of any agents for reducing moist desquamation from breast radiation based on a systemic review and meta-analysis of trials. But trials of skin agents, particularly older ones, are often of limited value because of their small numbers, variable dermatitis scoring systems used, end points that do not correlate with patient symptoms, or outdated radiation techniques. In addition, these agents often have no clear mechanism for reducing radiation dermatitis, so new agents with novel biologic rationale are needed.

The treatment of radiation dermatitis depends on its severity. There are substantial variations from one center to another in the preferred approaches and products used. Grade 1 to 2 dermatitis can be treated first with any number of water-soluble skin moisturizers such as MediChoice (Owens & Minor, Richmond, VA), Lotion Soft (STERIS, St. Louis, MO), or Eucerin (Beiersdorf, Hamburg). As symptoms of dryness, skin pain, or areas of dry peeling increase, petroleum-based emollients such as Aquaphor (Beiersdorf, Hamburg) can be used as needed. However, they should not be used immediately before treatment to prevent bolus effects that cause increased skin dose. Steroid creams may reduce mediators of inflammation in the skin that are responsible for radiation dermatitis, so a weak-strength hydrocortisone 1% to 2.5% or moderate-strength triamcinolone 0.1% may reduce papular eruptions and minor itching of the skin. Over-the-counter 1% to 5% lidocaine jelly can be used alone or mixed with Aquaphor for temporary relief of skin pain as well. Moist desquamation of the skin is managed by nonstick dressings, aluminum acetate solution in water (Domeboros, Moberg Pharma North America, Cedar Knolls, NJ), or silver sulfadiazine cream (Silvadene, King Pharmaceuticals, Bristol, TN). Prophylactic oral or topical antibiotics are not needed for moist desquamation.

Infections of the breast include fungal superinfections, particularly in the inframammary folds of large-breasted women or as a complication of moist desquamation. These may occur from overuse of antibiotics. Topical antifungal creams are applied for 10 to 14 days, with oral agents reserved for refractory cases. By the end of radiation, the presence of erythema and edema and skin pain can make detection of infection more difficult. Bacterial infections are associated with a marked change in skin color or pain, with or without fevers, unlike the usual more gradual progression of radiation dermatitis from week to week. A skin infection may have lesions that can be confused for papular radiation folliculitis ( Fig. 49.2 ). The pain or redness may appear out of proportion to the expected appearance from the number of treatments received when present earlier in a course of treatment. The incidence of developing cellulitis or breast abscess during or within 1 year of breast radiation is 1% to 8%. Cephalexin or ciprofloxacin is used for 10 to 14 days in cases of suspected superficial skin bacterial infection. For infections refractory to antibiotics, an ultrasound may be needed to rule out an abscess that requires incision and drainage.

Cosmesis and Breast-Related Symptoms

Potential long-term effects of radiation to the breast include breast pain, edema, fibrosis, induration, hyperpigmentation, or telangiectasias of the skin. Alone or combined these can have negative effects on the overall cosmetic result. Traditional scoring of breast cosmesis has been a simple scale of poor, fair, good, or excellent, with excellent representing no apparent effects of the radiation. Using this scale, approximately 80% to 95% of women have a good or excellent cosmetic results after postlumpectomy radiation. It is important to assess cosmesis at baseline before radiation, because the surgical effects may make affect cosmesis separately from radiation, in an attempt to distinguish what effects on cosmesis may be due to surgical changes rather than subsequent radiation changes. For example, changes in breast size can be immediate due to volume loss or delayed within months of surgery due to decrease in postsurgical edema or size of a seroma ( Fig. 49.3 ). In addition, patient-reported cosmetic results often are different than physician-reported or other objective measurements. For example, in one large prospective randomized study, patients reported approximately 94% good or excellent cosmesis compared with only 75% by a five-person panel evaluating their images at 5 years after surgery. There are many other scoring systems used. In some trials, cosmetic scoring may use terminology of mild, moderate, or severe normal tissue effects including edema, shrinkage, induration, or telangiectasia. For patient-reported outcomes, the Breast Cancer Treatment Outcome Scale (BCTOS) is a 22-item patient-reported measure of their esthetic cosmetic status, functional status, and breast-specific pain that has demonstrated validity.

Common symptoms after breast radiation include mild breast discomfort, sensitivity to touch, or shooting pains of the breast that come and go rapidly and unexpectedly. These will generally improve over time. In a randomized prospective clinical trial performed at Princess Margaret Hospital in Toronto, breast pain and quality of life were studied in women treated with BCS and tamoxifen with or without radiation. There was no significant difference between the arms in quality-of-life scores for physical function, pain, or breast symptoms within 12 months of treatment. In both treatment arms, the breast-reported symptoms and pain decreased over time, suggesting that the predominant factor causing them was surgery. Serious breast pain requiring medication occurs in approximately 1% of women. Initial management includes nonsteroidal antiinflammatory medication. More severe and persistent symptoms of the breast or chest wall may be treated with trials of medication such as low-dose nortriptyline, venlafaxine, or gabapentin.

Breast edema presents as heaviness, pain, or enlargement of the breast. Signs of edema include a global increase in breast size, skin thickening, or peau d’orange without other inflammatory signs. Skin thickening and increased breast density may also be apparent on the first postradiation mammogram. Delayed drainage from the nipple and areola may cause a distended appearance when moving from sitting to a supine position. Risk factors for breast edema are large breast size, axillary node dissection, and upper extremity lymphedema. Mild breast edema may be managed by wearing a sports bra that applies passive hydrostatic pressure to the breast. Moderate to severe cases of edema may be referred to physical therapy for therapeutic massage and decongestive therapy.

Skin telangiectasias occur in less than 10% of women treated with postoperative radiation, but there was higher incidence in older studies with use of cobalt or electron, a lumpectomy cavity boost, regional node irradiation, or concurrent chemotherapy. Telangiectasias or hyperpigmentation commonly develops within electron boost fields or skin folds due to higher skin dose in those locations ( Fig. 49.4 ). The time course to their appearance is usually 1 to 2 years, but they may continue to evolve for many years after. Management of asymptomatic telangiectasis is observation (benign radiation telangiectasias have no malignant potential), whereas in the rare case of symptomatic telangiectasis causing an unsightly appearance treatments used have included sclerotherapy, pulsed dye laser, or electrodessication.

Cosmetic outcomes in modern studies of whole breast hypofractionation have shown that use of a daily radiation dose higher than 2.5 Gy may no longer be associated with negative cosmetic results, and may actually have improved cosmetic outcomes when combined with sufficient reductions in the number of fractions and whole breast total dose. The Ontario Clinical Oncology Group trial randomized patients to 42.5 Gy in 16 fractions versus 50 Gy in 25 fractions without a tumor bed boost. The late cosmetic appearance was considered good or excellent in approximately 70% of women in both groups, with no reported differences in 10-year skin and subcutaneous tissue complications. The UK Standardisation of Breast Radiotherapy (START) trials consisted of two separate studies of whole-breast irradiation and hypofractionation that allowed a boost. Trial A compared 50 Gy in 25 fractions, 41.6 Gy in 13 fractions, or 39 Gy in 13 fractions in 5 weeks. Trial B compared 50 Gy in 25 fractions over 5 weeks versus 40 Gy in 15 fractions over 3 weeks. The late effects of breast appearance, breast edema or hardness, or skin changes were generally equal or better with whole breast hypofractionation. Other phase III trials have also confirmed that cosmetic outcomes are the same or better with whole-breast hypofractionation. There may be limits to hypofractionation—in a phase III trial of whole breast hypofractionation using only five large daily doses, there was noted to be a marked difference in normal tissue effects between just 26 Gy and 27 Gy were used to the whole breast with or without a boost.

The risk of late negative cosmetic effects after radiation may be increased by the use of a lumpectomy bed supplemental boost dose. In a large prospective randomized trial of dose escalation comparing 50-Gy whole-breast irradiation to 50 Gy plus a 16-Gy tumor bed boost, the use of the radiation boost was associated with an increased risk for severe fibrosis (1.8% in the no boost group vs. 5.2% in the boost group; P < 0.0001). In another randomized trial comparing 50 Gy in 25 fractions to 45 Gy in 25 fractions plus a 16-Gy boost, the physician and objective measurement of good or excellent cosmesis was approximately 80% with boost compared with 70% without boost. Other prospective trials did not observe a difference in cosmetic or soft tissue outcomes with use of a boost. These studies taken together suggest that other factors in addition to the use of a boost such as radiation technique, dose to the whole breast, large dose inhomogeneity, and total breast dose affect cosmesis. When the projected ipsilateral breast tumor recurrence (IBTR) reduction is low, a boost may not be regarded as medically required, and there are possible drawbacks such as decreased good or great cosmesis, increased soft tissue side effects such as fibrosis or telangiectasias, and an increased number of treatments and cost. In low-risk patients, including those over 70 years old with low- or intermediate-grade tumors and wide (2 mm) negative margins, an ASTRO task group recommended omitting the tumor bed boost. Concurrent chemotherapy is rarely used in modern clinical practice, but has been associated with increased rates of breast fibrosis, telangiectasia, hyperpigmentation, breast atrophy, and physician-reported worse cosmetic assemssements.

Measures to improve dose homogeneity may reduce the incidence of these complications. Modern computed tomography (CT)-based planning by either field in field or forward planned 3D conformal or inverse planned IMRT can reduce dose inhomogeneity and complications such as edema or negative cosmetic effects on the breast. In a trial conducted at the Royal Marsden Hospital in England, 306 women were randomized to receive whole-breast radiation after BCS using IMRT or conventional two-dimensional tangential radiation therapy. There was improved dose homogeneity with IMRT, with 19% of the IMRT plans showing dose inhomogeneity greater than or equal to 105% compared with 92% of the conventional plans. Photographic analysis 5 years after treatment showed that there was a change in breast appearance in 58% of patients randomized to conventional treatment, compared with 40% of those randomized to IMRT ( P = 0.008). There was a significant relationship between the presence of regions of the breast receiving greater than 105% of the dose and any change in breast appearance. Fewer patients treated with IMRT had clinically palpable induration of the breast tissue as well. In the Cambridge Breast IMRT trial, 1145 trial patients were analyzed, and 815 with a high predetermined level of dose inhomogeneity were randomized to standard wedged tangential radiation or replanned with a simple intensity modulation technique. The overall 5-year rates of moderate to good cosmesis were 88% with IMRT and 78% with controls and were significant on multivariate analysis. There was also a significantly lower incidence of skin telangiectasias with IMRT.

Underlying genetic factors that may affect radiation tolerance and soft tissue outcomes are not well known. BRCA mutation has not been associated with increased late radiation breast complications. An increase in late effects has been reported in patients with a heterozygous mutation of the gene for ataxia-telangiectasia (ATM) in many, but not all, studies. Ho and colleagues reported the increase in late grade 2 to 4 subcutaneous side effects was observed in particular association with a 5557G/A polymorphism of the ATM gene. There are no data on the effects of radiation complications by other breast cancer–associated incidence genes, but more data may be anticipated with the increase in the clinical use of commercially available multigene panels for testing newly diagnosed breast cancer patients.