Abstract
Background/Aim: To evaluate the safety and efficacy of lattice radiotherapy (LRT) for large, inoperable breast cancers. Patients and Methods: In this prospective study, patients who underwent LRT for breast tumors that were ulcerating/fungating/extensively eroding the chest wall, and were ineligible/unwilling for surgery, were enrolled from May 2021 to Nov 2023. Baseline Eastern Cooperative Oncology Group (ECOG) status, pre- and post-LRT numerical rating scale (NRS), and post-LRT changes in quality of life (QoL) were recorded. Survival outcomes were reported at 6 months and 1-year. Median rates of survival and dosimetric parameters were calculated. Kaplan–Meier curves for overall survival (OS), cancer-specific survival (CSS), and failure of local control (LC) were constructed. Results: Ten patients (8 females) underwent LRT. The median age was 76 years (range=57-99 years) and the median ECOG performance status was 2.5 (range=1-4). The planned schedule was completed by 9/10 patients, accounting for a 90% compliance rate. Among patients with pain (n=7), NRS rapidly reduced from 7 (range=5-10) to 3 (range=1-6). The median equivalent uniform dose was 0.71 Gy (0.09-1.59 Gy). The actuarial rates of 6-month LC, CSS, and OS were 75%, 89%, and 61%, respectively, with only LC rate changing to 50% at 1 year. Two patients had local relapse at the six-month and 1-year follow-up, respectively, after having achieved a complete response at three months, and two others died of COVID-19 infection and ischemic stroke. Conclusion: LRT was found to be effective and safe in palliating symptoms among patients with large inoperable breast tumors.
- Lattice radiotherapy
- breast cancer
- radiation oncology
- palliative therapy
- fungating breast cancer
- ulcerating tumor
Newly diagnosed large primary breast cancers with extensive skin and/or chest wall involvement may particularly affect patients with psychosocial issues (i.e., psychiatric disorders or financial and logistical matters), which hinder or delay access to healthcare (1, 2). The occurrence of skin ulceration, bleeding, foul smell, or pain requires timely intervention to reduce the risk of superinfection and anemia, and also serves to avoid further deterioration of quality of life (QoL) (3, 4). In these cases, given the inadequacy of systemic therapies (i.e., chemo-, hormonal, and immunotherapy) in producing a rapid tumor response, and considering that electrochemotherapy is not yet widely available, palliative radiotherapy (RT) and mastectomy are often the only viable therapeutic options (5-7). However, surgery may not be indicated due to the patient’s critical clinical condition, patient refusal, or technical non-feasibility following the anticipated inability to achieve clinically negative margins for effective wound healing. In this context, RT is an attractive non-invasive treatment.
Although classic palliative RT schedules (i.e., 20-30 Gy in 5-10 daily fractions) can achieve pain relief and hemostasis (8), they may not suffice when lasting cytoreduction of a fungating mass is required to effectively restore skin integrity. Indeed, this has encouraged the testing of moderately escalated radiation doses (9). On the other hand, stereotactic body radiotherapy using doses in the ‘ablative’ range can result in a significant shrinkage or even disappearance of early-stage breast cancers (10). Such high doses may not be suitable for homogeneous irradiation of bulky tumors due to unfavorable dose-volume effects, which threaten the radiation tolerance of nearby organs at risk (OARs) (11). Therefore, in the setting of very large breast cancers, LATTICE radiotherapy (LRT), comprising spatially alternating high and low doses, may emerge as a convenient option. LRT optimally balances the need for tumor downsizing and symptom palliation without exceeding the tolerance of OARs, as already proven in some mixed cohorts (12). LRT is a feasible way to safely escalate the radiation dose within some tumor subvolumes.. The peculiar spatial dose pattern can trigger specific anti-tumor immune responses by simultaneously employing radiation doses ranging from very low (valleys) to very high (peaks) (13).
Herein, we present the first series of patients with large or inoperable breast cancer who underwent LRT, reporting the technical issues, patient compliance, palliation of symptoms, toxicity, and survival outcomes.
Patients and Methods
Study design and setting. This prospective observational study was conducted at a tertiary care hospital in Viagrande, Italy from May 2021 to Dec 2023. Since May 2021, our center has been offering LRT to patients who are ineligible or unwilling for surgery of large, inoperable breast cancers.
Patient enrolment. As per the inclusion criteria, patients (i) with breast cancers that were ulcerating or fungating through the skin or extensively eroding the chest wall and (ii) were ineligible or unwilling for surgery were enrolled, regardless of systemic staging (metastatic or not). Exclusion criteria included superinfection with fever, non-compliance with setup procedures, and patient refusal.
The typical LRT candidate complained of a feeling of heaviness (either due to the tumor mass or swelling following lymphatic obstruction), burning, pain with or without arm lymphedema (due to the dramatic distension of the soft tissues by the bulky primary tumor, infiltration of the intercostal nerves, or compression of the axillary vascular-nervous bundle by large lymphadenopathies), or frequent dressing changes (due to foul-smelling exudate or bleeding from the malignant wound).
Treatment. All patients required histological confirmation of breast cancer prior to LRT. Pre-treatment instrumental staging was required but not mandatory, as it did not modify the therapeutic path. It was generally performed using total body contrast-enhanced computed tomography (CT); otherwise, the simulation CT scan was extended from the neck to the abdomen for minimum staging evaluation of patients with poor compliance. In cases of disruption of skin integrity with seepage, malodorous discharge, or bleeding, patients were supported by professional nursing care (14, 15).
LRT program. Once a simulation CT scan with 1.25 mm thickness slices was acquired in the supine position with arms above the head over a breast board, two plans were generated for a sequential treatment: 1) an anticipated boost by an LRT_plan followed by 2) a palliative hypofractionated RT (hypoRT_plan). The use of a bolus to make the radiation dose also cover the skin surface was optional, since the treatment delivery by volumetric modulated arc therapy (VMAT) may somewhat address this need taking advantage of the tangential effect of the rotational RT techniques (16). The target contours were delineated as follows: the gross tumor volume (GTV) was represented by the macroscopic primary tumor and included in the clinical target volume (CTV), which encompassed the remaining breast tissues up to the skin surface and draining lymph node basins. The CTV was expanded by 0.5 cm to the planning target volume (PTV). A variable number of 1-1.5 cm vertices (Vx) was randomly delineated within the GTV. No minimum and maximum number of vertices was set, a distance of at least 4-5 cm between them (from center to center) being the only prerequisite. The lungs, heart, spinal cord, and healthy chest wall were contoured as OARs. The LRT_plan simultaneously delivered a single fraction of stereotactic high radiation dose to the vertices alone (Vx_sum). After a short rest, i.e., 3-4 days, a palliative hypoRT_plan, consisting of 5-15 daily fractions, targeted the whole PTV as determined above. Both plans were delivered by a Varian Truebeam Novalis STx using VMAT. The 98% of the prescription dose of the LRT_plan and hypoRT_plan had to cover the 100% volume of each vertex and the 98% of the PTV, respectively. Moreover, as regards the LRT_plan, a few other features were annotated, such as the vertices-GTV volume ratio (Vx_sum/GTV), the peak dose prescribed to the vertices (Dp), the maximum vertex dose (Dmax), and the lowest dose within the GTV (valley dose). By the dose-volume histogram (DVH) of the LRT_plan, the GTV equivalent uniform dose (EUD) was determined according to Niemierko, while assuming that the corresponding one in the hypoRT_plan was almost equal to the prescribed dose since this was delivered as homogeneously as possible to the whole PTV (17). The valley-to-peak dose ratio (VPDR) was calculated according to Wu et al. in the LRT_plan (18), taking into account that the LATTICE volume was equivalent to the GTV since the whole breast and overlying skin were already included in the CTV and no inward safe margin was necessary when judiciously avoiding high-dose vertex placement away from the ribs and intercostal nerves (i.e., 2-3 cm apart), these being the sole organs at risk potentially abutting the GTV. Given the fact that the lack of a prespecified geometric arrangement of vertices could affect the VPDR, a further parameter was used to evaluate the LRT_plan quality, i.e., the Paddick Conformity Index (PCI) (19). This served to assess the conformity of the prescribed dose to the vertices, as low values (tending to 0) may indicate a “bridge” high dose between them. A plan_sum combining the LRT_plan and hypoRT_plan was created to assess the cumulative dose to OARs. The dosimetric goal for the chest wall was to not exceed a combined maximum EQD2 of 60 Gy (Equivalent Dose in 2 Gy Fractions using an α/β value of 3 Gy for chest wall). The combined mean dose to the heart was constrained within 4 Gy (Dmean_heart <4 Gy), while the relative volume of the lungs receiving a combined dose higher than 2.5 Gy was kept below 50% (V2.5Gy_lungs <50%). Given the limited life expectancy, no special considerations were made to spare the left anterior descending coronary artery in patients with left breast cancer (20, 21).
Each RT fraction was verified daily, according to our image-guided protocol using a cone-beam CT, which enabled us to assess any volumetric changes in the irradiated tumor throughout the RT schedule, in addition to helping correct positional errors prior to radiation delivery. Finally, the time elapsed between the end of RT and the start of systemic therapy, if any, was also recorded.
Outcomes. Baseline performance status was graded according to the Eastern Cooperative Oncology Group (ECOG). Pre- and post-LRT pain assessment was scored using a numeric rating scale (NRS) ranging from 0 (no pain at all) to 10 (pain as bad as it could be) (22). Patients were also interviewed about their QoL, recording any improvement or worsening after LRT. Treatment toxicity was evaluated according to the Radiation Therapy Oncology Group (RTOG) criteria and the Common Terminology Criteria for Adverse Events (CTCAE) version 4.1. Patients were monitored during the LRT schedule and submitted to clinical examination one month after completion of treatment and then also to CT follow-up every three months. Such imaging permitted the assessment of any volumetric changes in the irradiated tumor according to the Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1, defined as follows: progressive disease (PD), stable disease (SD), partial response (PR), and complete response (CR), the latter two being combined in the objective response rate (ORR) (23). Confirmation by 18F-Fluorodeoxyglucose positron emission tomography was required based on clinical judgement in ambiguous cases of CR, to differentiate between residual scar/fibrosis and disease persistence. Although the advanced stage breast cancers presented here indicate palliation of symptoms as the primary goal of LRT, the following survival outcomes were also recorded from the end of LRT: local control (LC), cancer-specific survival (CSS), and overall survival (OS).
Ethical statement. This prospective observational study was conducted in accordance with the Declaration of Helsinki, and approved by Messina Ethics Committee of AOU Policlinico “G. Martino” with protocol code 1611-38-21. Written informed consent was obtained from all patients involved in the study for the publication of this manuscript and all accompanying images.
Statistical analysis. The median values of dosimetric parameters, survival, and local control were calculated using Microsoft Excel. The survival functions for OS, CSS, and failure of local control were plotted using the Kaplan–Meier method (24) using IBM® SPSS® software version 25 (IBM, Armonk, NY, USA). The manuscript was written in accordance with the STROBE guidelines for observational studies.
Results
From May 2021 to November 2023, after multidisciplinary consultations, we enrolled 10 patients, including 8 females, for LRT. Seven patients had a left-sided tumor. Demographics and clinical characteristics are depicted in Table I. The median age was 76 years (range=57-99 years) and the median ECOG performance status was 2.5 (range=1-4). A fungating tumor was present in five patients; three others had a bulky tumor with the overlying skin being red, hot, and tense; the remaining two had several crusty lumps tending to bleed in the right breast and a painful post-mastectomy recurrence eroding the full-thickness of the left chest wall, respectively. All patients were judged inoperable due to technical issues, significant comorbidities, or advanced age. The main signs and symptoms were bleeding (n=2), malodor (n=3), wound exudate (n=6), arm lymphedema (n=1), breast swelling and tenderness (n=7), and pain (n=7).
Demographics and clinical characteristics of patients.
The planned schedule was completed by nine out of ten patients, accounting for a 90% compliance rate, since one 91-year-old female refused to continue after delivery of the LRT_plan because of excessive discomfort in lying still during the treatment and was subsequently lost to follow-up. Early successful palliation was achieved in all patients treated, regardless of the symptoms they complained of. Among patients with pain (n=7), the NRS score was rapidly reduced from 7 (range=5-10) to 3 (range=1-6). All patients who completed the RT program perceived a significant improvement in their QoL. Eight/nine patients started a systemic therapy after a median time of 15 days from the end of RT.
Dosimetric parameters of all LRT_plans are shown in Table II, together with the dose prescription of subsequent hypoRT_plans (last column on the right). The median EUD of LRT_plans was 0.71 Gy (range=0.09-1.59 Gy). Regarding the post-RT volumetric changes, we observed 4 CR and 3 PR, accounting for a 100% ORR, at three-month follow-up. Significant downsizing of GTV was evident already during the RT program in 4/9 patients, long before the start of systemic therapy, if any.
Dosimetric report of LATTICE radiotherapy plans for the ten patients enrolled in this study.
With a median follow-up of 3.8 months, the median was not reached for any of the survival outcomes. The actuarial rates of 6-month LC, CSS, and OS were 75%, 89%, and 61%, respectively, with LC being the only survival outcome to change to 50% at 1 year. Two patients had local relapse at the six-month and 1-year follow-up, respectively, after having achieved a CR at three months, and two other patients died of COVID-19 infection and ischemic stroke. We recorded only one case of CTCAE grade 2 radiodermatitis as the worst toxicity. Two illustrative cases of LRT for a fungating bulky breast tumor are represented in Figure 1 and Figure 2, whereas survival curves are summarized in Figure 3.
An illustrative case of a fungating left-sided breast cancer (Patient 2 in Tables I and II) (A) prior to radiotherapy (RT), (B) during RT, (c) one, (D) three, and (E) six months after RT completion; (F) dose color wash for the LRT plan, where 98% of the dose prescribed to three vertices is visible in blue/green; (G) dose color wash for the plan sum (LRT_plan + hypoRT), where the 95% of the dose prescribed to the PTV fits its contour, within which the distribution of highest doses is conformed to the gross tumor volume (GTV) and vertices; (H) dose-volume histogram (DVH) for the LRT_plan: steep red lines are for all 14 vertices, while the red one that slopes gradually down is for GTV; yellow line is for the heart, green one for the lungs; (I) DVH for the plan sum (LRT_plan + hypoRT_plan), where the red line on the left is for the PTV, while the red one on the right is for the vertices_sum volume; yellow line is for the heart, orange line is for the left anterior descending coronary artery and the green line is for the lungs.
An illustrative case of a fungating right-sided breast cancer (Patient 7 in Table I and Table II) (A) prior to radiotherapy (RT), (B) first follow-up 1 month after RT (complete response). (C) Dose color wash for the LRT plan, where 98% of the dose prescribed to two vertices is visible in blue/green; (D) dose color wash for the plan sum (LRT_plan + hypoRT), where the 95% of the dose prescribed to the PTV fits its contour, within which the distribution of highest doses is conformed to the gross tumor volume (GTV) and vertices; (E) dose-volume histogram (DVH) for the LRT_plan: steep red lines are for all 6 vertices, while the red one that slopes gradually down is for GTV; yellow line is for the heart, green one for the lungs; (F) DVH for the plan sum (LRT_plan + hypoRT_plan), where the red line on the left is for the PTV, while the red one on the right is for the vertices_sum volume; yellow line is for the heart and the green line is for the lungs.
Kaplan–Meier survival curves depicting the duration of local control (below), overall survival (at the top left), and cancer-specific survival (at the top right) in months.
Discussion
To our knowledge, this is the first clinical report specifically devoted to the LRT treatment of primary breast cancers, being preceded only by a dosimetric technical note (25). Except for a single case of tumor lysis syndrome (26), LRT has been shown to be safe and effective in several mixed cohorts and case reports, none of which have included breast cancers (13, 17, 27, 28). It has mainly been used to downsize inoperable bulky tumors, which were not amenable to homogeneous irradiation with stereotactic high doses due to tolerance issues.
Indeed, the spatial dose pattern of LRT allows an ablative dose to be delivered in at least some inner subvolumes of the tumor mass, without exceeding the tolerance of the peritumoral OARs (12). The resulting overall tumor shrinkage may not fit the response predicted by the linear quadratic model and rather follow the activation of anti-tumor immune mechanisms (29). In our cohort, the median EUD of LRT_plans resulting from low VPDR values was low as well (0.71 Gy), thus not adding a significant contribution to the palliative potential of subsequent hypoRT_plans. Contrary to expectations, we observed a significant volume reduction in 5/9 tumors already during the RT course, supporting the hypothesis of an early immune intervention that prevails over the direct cytotoxic effect of radiation (30). In this view, compared to the LRT technique using a simultaneous integrated boost (31), our sequential approach with no dose prescribed to the tumor periphery in the LRT_plan has the advantage of sparing some tumor vessels and circulating lymphocytes (32). These immune cells, known to be highly radiosensitive (33), can be reprogrammed against the tumor antigens released from the high-dose subvolumes in the pause before the hypoRT_plan delivery (34). This short pause would allow an effective priming and expansion of lymphocytes, whose repopulation guided by immune memory may compensate for their intratumoral depletion subsequent to the hypoRT_plan delivery.
Moreover, compared to a rigidly geometric arrangement of vertices, a random one is more versatile as it ensures feasibility by simply avoiding vertex delineation near the OARs. This ensured that we did not exceed the tolerance dose of OARs, i.e., chest wall, heart, and lungs. Accordingly, we recorded no toxicity greater than that expected from classic adjuvant RT to the breast. The only treatment interruption was reported due to non-compliance with the setup procedures. All other patients, including the elderly, completed the planned schedule, reflecting an excellent compliance rate (90%). All treated patients benefitted from LRT, having achieved effective palliation of their symptoms. They also reported a significant improvement in their QoL. Surprisingly, some patients became long survivors in good health. It is possible that the baseline poor prognosis was reversed by the subsequent systemic therapies, probably facilitated in their success by an immunostimulatory effect of the upfront RT (35, 36). Systemic therapy alone, in fact, has proven to be ineffective in controlling such advanced tumor masses that, on the contrary, can regress once RT is introduced (37). Only two local failures were recorded within the first year after RT, both asymptomatic and without worsening of QoL.
From this brief report, LRT emerges as a convenient and viable therapeutic option for patients with large breast tumors, functioning as an attractive alternative to surgery, which, in contrast, is fraught with serious risks, especially among frail and elderly patients in whom death may occur from causes unrelated to cancer. In this scenario, mastectomy is burdened by high morbidity, as proven by Abdallah et al. whose large cohort experienced a complication rate of 40.7%, of which almost half consisted of the Clavien-Dindo grade II category. Interestingly, the local failure rate (20.7%) was comparable to that reported by us with a non-invasive treatment (38). As for the definition of the best RT regimen, the homogeneous hypo-fractionated ones used by Chatterjee et al. revealed an overall palliation success rate worse than ours at the 3-month evaluation, although their dose prescriptions were similar to those used in our hypoRT_plans and followed pre-treatment with systemic therapy in most patients (9). However, both samples are not large enough to draw definitive conclusions. We agree with these authors on the need to shorten the overall treatment time by adopting hypo-fractionated schedules to limit patient stress. An anticipated boost by LRT does not excessively prolong the overall treatment time, thus not threatening the patient’s compliance with the subsequent hypo-fractionated RT schedule.
Study limitations. First, the limited sample size reflects a very low recruitment rate, i.e., 10 patients in nearly three years. This could be justified by the rarity of the advanced disease stage investigated here, especially in a developed country with a capillary network of healthcare facilities, where advanced clinical presentation occurs mainly due to patients’ fears, emotional distress, and reluctance to see a physician (7). Second, different histologies, as in our report, can lead to different outcomes, which may be strictly dependent on the efficacy of specific systemic therapies (39). However, at least palliation and tumor reduction to an extent were successfully achieved even before any role for systemic therapies could be adduced. Third, vertex delineation was performed at the discretion of the treating physician. While this improves versatility and safety, it also impairs the reproducibility of results. Indeed, our dosimetric findings (i.e., VPDR and EUD), as well as the volumetric parameters behind them (Vx number and Vx_sum/GTV in the absence of a predetermined template for the arrangement of vertices), were collected post hoc and not entered into the inverse planning algorithm for optimizing the LRT_plan. A physician-supervised automated treatment planning approach could compensate for such a need in the next future (40).
Finally, we promote the integration of multidisciplinary strategies to effectively manage these challenging clinical pictures, which should be faced by specialized care teams, hopefully within breast units (41, 42).
Conclusion
LRT was found to be effective and safe in palliating symptoms and improving the QoL among patients with large inoperable breast tumors. It could potentially emerge as a treatment for prolonging survival as well. Larger clinical reports and preclinical studies aimed at elucidating the underlying molecular mechanisms are needed to confirm these preliminary results and possibly exploit the full potential of LRT.
Footnotes
Authors’ Contributions
Conceptualization: Gianluca Ferini; Methodology: Valentina Zagardo, Gianluca Ferini; Formal analysis: Mandara Harikar; Writing – original draft: Gianluca Ferini; Writing – Review & Editing: Gianluca Ferini, Valentina Zagardo, Mandara Harikar, Anna Viola, Domenico Patanè, Silvana Parisi, Paolo Fontana, Giovanni Maugeri, Angela Prestifilippo, Antonio Piras, Francesco Cuccia, Andrea Boncoraglio, Antonio Pontoriero; Supervision: Stefano Pergolizzi.
Conflicts of Interest
The Authors have no conflicts of interest to declare in relation to this study.
Funding
This study received no external funding.
- Received May 30, 2024.
- Revision received June 24, 2024.
- Accepted June 26, 2024.
- Copyright © 2024 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.