Palliative radiotherapy indications during the COVID-19 pandemic and in future complex logistic settings: the NORMALITY model

Introduction The COVID-19 pandemic has challenged healthcare systems worldwide over the last few months, and it continues to do so. Although some restrictions are being removed, it is not certain when the pandemic is going to be definitively over. Pandemics can be seen as a highly complex logistic scenario. From this perspective, some of the indications provided for palliative radiotherapy (PRT) during the COVID-19 pandemic could be maintained in the future in settings that limit the possibility of patients achieving symptom relief by radiotherapy. This paper has two aims: (1) to provide a summary of the indications for PRT during the COVID-19 pandemic; since some indications can differ slightly, and to avoid any possible contradictions, an expert panel composed of the Italian Association of Radiotherapy and Clinical Oncology (AIRO) and the Palliative Care and Supportive Therapies Working Group (AIRO-palliative) voted by consensus on the summary; (2) to introduce a clinical care model for PRT [endorsed by AIRO and by a spontaneous Italian collaborative network for PRT named “La Rete del Sollievo” (“The Net of Relief”)]. The proposed model, denoted “No cOmpRoMise on quality of life by pALliative radiotherapy” (NORMALITY), is based on an AIRO-palliative consensus-based list of clinical indications for PRT and on practical suggestions regarding the management of patients potentially suitable for PRT but dealing with highly complex logistics scenarios (similar to the ongoing logistics limits due to COVID-19). Material and Methods First, a summary of the available literature guidelines for PRT published during the COVID-19 pandemic was prepared. A systematic literature search based on the PRISMA approach was performed to retrieve the available literature reporting guideline indications fully or partially focused on PRT. Tables reporting each addressed clinical presentation and respective literature indications were prepared and distributed into two main groups: palliative emergencies and palliative non-emergencies. These summaries were voted in by consensus by selected members of the AIRO and AIRO-palliative panels. Second, based on the summary for palliative indications during the COVID-19 pandemic, a clinical care model to facilitate recruitment and delivery of PRT to patients in complex logistic scenarios was proposed. The summary tables were critically integrated and shuffled according to clinical presentations and then voted on in a second consensus round. Along with the adapted guideline indications, some methods of performing the first triage of patients and facilitating a teleconsultation preliminary to the first in-person visit were developed. Results After the revision of 161 documents, 13 papers were selected for analysis. From the papers, 19 clinical presentation items were collected; in total, 61 question items were extracted and voted on (i.e., for each presentation, more than one indication was provided from the literature). Two tables summarizing the PRT indications during the COVID-19 pandemic available from the literature (PRT COVID-19 summary tables) were developed: palliative emergencies and palliative non-emergencies. The consensus of the vote by the AIRO panel for the PRT COVID-19 summary was reached. The PRT COVID-19 summary tables for palliative emergencies and palliative non-emergencies were adapted for clinical presentations possibly associated with patients in complex clinical scenarios other than the COVID-19 pandemic. The two new indication tables (i.e., “Normality model of PRT indications”) for both palliative emergencies and palliative non-emergencies were voted on in a second consensus round. The consensus rate was reached and strong. Written forms facilitating two levels of teleconsultation (triage and remote visits) were also developed, both in English and in Italian, to evaluate the patients for possible indications for PRT before scheduling clinical visits. Conclusion We provide a comprehensive summary of the literature guideline indications for PRT during COVID-19 pandemic. We also propose a clinical care model including clinical indications and written forms facilitating two levels of teleconsultation (triage and remote visits) to evaluate the patients for indications of PRT before scheduling clinical visits. The normality model could facilitate the provision of PRT to patients in future complex logistic scenarios.


Introduction
The COVID-19 pandemic has challenged healthcare systems worldwide over the last few months, and this is still ongoing [1,2]. Although some restrictions have been removed depending on the country, it is not certain when the pandemic is going to be over for certain. Radiation oncologists (ROs) will be forced to face the pandemic for an unknown time interval. Two main approaches have been adopted to control COVID-19: suppression and mitigation, the latter being the most frequently adopted [3]. The mitigation approach imposes the need for indications for how and when to delay or omit radiotherapy (RT). Alternatively, hypofractionated RT schedules, which adequately manage different clinical settings, have been proposed to reduce the number of interactions and physical contact in hospitals (for both patients and patients) while delivering effective treatments [4][5][6][7].
National and international guidelines and expert opinions about RT indications and prescriptions have been provided for primary malignancies (e.g., head and neck [6] or gastrointestinal [7]). More often, palliative RT (PRT) indications during the COVID-19 pandemic scenario are dealt with using reports focused on primary malignancies. To the best of our knowledge, very few guidelines have been specifically dedicated to PRT, and in some cases, these are limited to particularly relevant palliative presentations (e.g., bone metastases [8]). Although the level of priority of PRT has frequently been the object of discussion [3,[9][10][11], it remains one of the primary aims of RT. Once the COVID-19 pandemic has concluded, many of the RT indications currently modified due to pandemic issues could not be further considered for most primary tumors. Conversely, in some situations (e.g., patients admitted in hospice; patients living at high distance from an RT department; less-resourced developing countries), the issue of patients suitable for PRT but dealing with complex logistic settings and thus subject to limitations in their possibility to achieve symptom relief by PRT will surely persist. Therefore, some of the indications provided for PRT during the COVID-19 pandemic could be safely and effectively maintained in these peculiar settings (since they are currently clinically accepted). This paper has two aims: (1) to provide a summary of the indications for PRT during the COVID-19 period. Since some published guidelines are slightly different, in order to harmonize the suggestions, an expert panel composed of the Italian Association of Radiotherapy and Clinical Oncology (AIRO) and the Palliative Care and Supportive Therapies Working Group (AIRO-palliative) voted by consensus on the summary. (2) To introduce a clinical care model for PRT [endorsed by AIRO and by a spontaneous Italian collaborative network for PRT named "La Rete del Sollievo" ("The Net of Relief")]. The proposed model, denoted "No cOmpRoMise on quality of life by pALliative radIoTher-apY" (NORMALITY), is based on an AIRO-palliative consensus-based list of clinical indications for PRT and on practical suggestions regarding the management of patients potentially suitable for PRT but dealing with highly complex logistics scenarios (similar to the ongoing logistics limits due to .

Material and methods
The two aims of this project were handled separately and progressively. The first aim (1) was to summarize the PRT clinical indications during the COVID-19 pandemic from the available RT literature. In particular, we aimed to (1.1) provide a summary (PRT COVID-19 summary) of the indications and guidelines for PRT during the COVID-19 period and to (1.2) have the AIRO expert team group vote by consensus on the PRT COVID-19 summary.
Our second (2) aim was to create the NORMALITY clinical care model ("No cOmpRoMise on quality of life by pALliative radIoTherapY"). In particular, we aimed to (2.1) provide a set of PRT indications for patients dealing with complex logistic scenarios (strongly limiting their possibility of receiving PRT beyond its given clinical indication) in order to (2.2) provide practical advice and supportive materials to optimize the clinical management of these patients by RT departments.

Create the PRT COVID-19 summary
A systematic literature search based on the PRISMA approach was performed by two radiation oncologists (ROs; RDF and VB). The search was performed using PubMed. We applied the following medical subject headings (MeSH) and keywords such as: "Radiotherapy," "Radiation Therapy," "Radiation Oncology," "Palliative," "Palliative Radiotherapy," "COVID-19," and "SARS-COV2." The detailed Medline search strategy is reported in Appendix.
For the first literature search, other documents were added by a manual search performed by a third RO (SM). The review was strictly composed of full-text publications that were written in English and reported clinical indications for PRT to be applied during the period of the COVID-19 pandemic. The literature search was conducted on April 26, 2020. All types of publications were initially considered, including surveys, letters, and editorials, provided that the prescriptive indications for PRT were clearly reported. Papers reviewing literature or personal considerations and not directly addressing a prescriptive indication were excluded. Reports of congress abstracts and book chapters were excluded. Reports providing clinical PRT indications and those that did not undergo a peer-review process were also excluded. No specific time restrictions other than those implicitly related to the COVID-19 pandemic period were applied. An independent literature revision was made by a different RO who supervised the summary consolidation process (FC). Eligible citations were retrieved for full-text review. Figure 1 illustrates the PRISMA workflow.
To homogenize the summary output, the clinical presentations discussed in the documents were collected into two main groups: emergencies and palliative non-emergencies.

3
A retrospective collection of each clinical presentation was organized into these two groups within multiple clinical presentation subgroups (defined as "clinical presentation items," CPI). Three information categories were extracted from each selected document: (i) the main (preferable) PRT prescriptive indication; (ii) the alternative (secondary) PRT prescriptive indication; (iii) additional statements in the document per each subgroup clinical presentation not strictly indicating a PRT prescription but specifically aimed at the considered subgroup topic.
A document was excluded by a certain clinical presentation subgroup if none of the three categories were addressed, but possibly included for other subgroup topic indications.
A group of four separate ROs double-checked the different sections of the PRT COVID-19 summary to confirm the correspondence of the data extraction (FA, AT, GS, and AC).

AIRO expert team consensus vote on the PRT COVID-19 summary
An expert panel of 14 AIRO members, which was not involved in any of the previously described phases, was asked to vote on, in a single round, the consensus of each of the reported PRT indications. Consensus was addressed by four options: 1 = "strongly agree," 2 = "agree," 3 = "disagree," and 4 = "strongly disagree". Consensus was based on all of the indications from each paper (i.e., main + alternative PRT indication + additional statements), thus preventing experts from only agreeing or disagreeing with specific parts of the summarized papers.
For each paper reported in the summary, the results were analyzed by a single vote and coupled as either a "positive consensus vote" (i.e.: 1 + 2) or "negative consensus vote" (i.e.: 3 + 4).
As for other experiences [5], an agreement or disagreement threshold ≥ 66% was required for each item to reach a consensus and a threshold of ≥ 80% was required for a strong consensus.

NORMALITY ("No cOmpRoMise on quality of life by pALliative radIoTherapY") clinical care model
The spontaneous network named "La Rete del Sollievo" (www. laret edels ollie vo. net) (i.e. Net of Relief, NOR), which is set up at the Department of Radiation Oncology of Fondazione Policlinico A. Gemelli IRRCS (Rome, Italy), promotes palliative radiation oncology clinical care models and shares research projects in collaboration with the AIROpalliative panel. Under the endorsement of the AIRO, the NOS aimed to create a clinical care model for patients with an indication for PRT who are dealing with high complexity logistic scenarios that limit their possibility of receiving a regular PRT schedule.

Provide PRT indications (through consensus vote by the AIRO expert team)
From the evidence base of the PRT COVID-19 summary, a table of PRT indications for patients in complex logistic settings (potentially other than the COVID-19 pandemic) was set up for the NORMALITY clinical care model. The table "Normality model of PRT indications" followed the structure of the PRT COVID-19 summary. The AIRO expert panel voted in two rounds of consensus using the same previously described methodology. In this case, we also added the opportunity to provide comments and alternative indications for the first voting round only. After the revision of the first-round votes and comments, a final version of the table was voted on once more. Analysis of the consensus was performed as previously described. Only the results of the final consensus round were considered in the analysis.

Provide practical advice and supportive materials for the NORMALITY clinical care model
To establish the practicalities of the NORMALITY clinical care model, the definition of the workflow was addressed based on the available literature indications [9,12], some currently ongoing practices among the Radiation Oncology departments of the AIRO experts involved in the project, and through discussion among the AIRO experts. The core concept was the advantage of a preliminary evaluation of the patient's indications ahead of a live visit. During the live visit, the PRT prescription would be confirmed, potentially including (within the same day) the RT simulation and the first (or single) session administration. Moreover, two types of forms aiding practical patient management were prepared: one to perform the first general patient data collection and allow for triage in palliative settings, and the second to aid the offline evaluation of patients ahead of clinical visits.
In total, 61 question items to be voted on were extracted from the papers including the 19 CPIs. Table 1 presents the PRT COVID-19 summary for palliative emergencies. Table 2 presents the PRT COVID-19 summary for palliative non-emergencies, along with the consensus results. References cited by selected papers are also reported in the table, if specifically related to trials [5,. The average agreement was over the agreement threshold, with only 10/61 question items from the different evidence having an agreement below 60% and five that were below 50%. The latter was related to single evidence regarding painful bone metastases, bone metastases suitable for SBRT, adjuvant treatment of bone metastases, other than pain symptoms related to lung primary, and SBRT for adrenal lesions.

NORMALITY ("No cOmpRoMise on quality of life by pALliative radIoTherapY") clinical care model
The NORMALITY clinical care model aims to make the stays in RT departments of patients dealing with complex logistic settings (e.g., in home care or hospice, living a long distance from the closest RT department) as short as possible. Ideally, patients should receive clinical visits, simulations, and single (or first) PRT delivery on the same day. Single-fraction PRT should be preferred whenever possible, unless the risk of unacceptable toxicity cannot be avoided. This was realized in some fast-track or rapid-response RT programs [12,[51][52][53]. The integration proposed to such acknowledged care models is to prepare ahead of patient arrival via least two levels of teleconsultation (triage and remote visits). The first level (triage) aims to enable the triage of patients possibly requiring PRT through a simplified information collection method that can be performed by a clinician or a qualified nurse. The triage can subsequently require a remote visit. This second-level contact with the patient (remote visit) involves a single or repeated remote visit, with potentially more in-depth information collected by the RO in order to administer the PRT prescription. If imaging evaluation is needed, the caregiver can be asked to acquire imaging, or alternatively (depending privacy rules), sharing through a computer network could be considered. Teleconsultation for triage and remote visits can be done by interactive and video calls, but also through phone calls that can effectively respond to such needs [12]. Figure 2 represents a form in the English version that facilitates the first triage (the Italian version is shown in Fig. 3). Figure 4 represents a form (in the English version) to facilitate remote visits (the Italian version is shown in Fig. 5). The summary of the PRT indications (i.e.: PRT Normality model summary) for this peculiar setting (not belonging to regular PRT) with the relative consensus is reported in Table 3 for palliative emergencies and Table 4 for palliative non-emergencies. With respect to the PRT COVID-19 summary, in PRT Normality model summary 20 CPIs were identified and one additional subtype clinical presentation was included: non-painful bone metastases. Thirty subtype indications for the 20 CPIs were summarized. The average agreement ("agree" + "strongly agree") was over the strong agreement threshold (i.e.: 80%), ranging from 82 to 100% among the 30 topic items, of which the inner rate of agreement of the first vote ranged from 33 to 92%.
The latest versions of such materials can also be retrieved in the following section of the "La Rete del Sollievo" (NOS) website (http:// www. gemel liart. it/ laret edels ollie vo/ reted elsol lievo-model liass isten ziali/). An interactive list of Italian RT departments providing different palliative services endorsed by the AIRO can also be downloaded from this website.

Discussion
Our paper aims to deal with three main issues regarding PRT: how to choose a PRT prescription during the COVID-19 pandemic; highlight the priority of administering PRT for patients both during the COVID-19 pandemic and in the future; and how to manage the risk of underuse of PRT in the future in patients dealing with complex logistic scenarios (particularly after the emergency pandemic experience, which suggests that different approaches to PRT are preferable if the RT department is inaccessible).
How to choose a PRT prescription during the COVID-19 pandemic? The COVID-19 pandemic challenges healthcare systems worldwide [1,54]. Some authors described that two phases are possibly expected from an RO's perspective: an early phase in which the department's human resources are not limited and potentially all treatments could be provided for patients; and a second phase ("late phase or scenario") in which the infection spread could limit the global amount of actually deliverable RT treatments [5]. To address this scenario, hypofractionated RT schedules have been proposed

P14 Brain metastases (N° 5-10)
QP14a Simcock [14] Palliation WBRT:     • For highly selected clinical settings of extensive bone involvement, or retreatment/ pain refractory to pain killers: cautiously consider "Half-body RT" (i.e.: lumbar + bony pelvis + femurs-15 Gy/4fx3.75 Gy BID) [SHARON project [35]]        to reduce the number of contacts while effectively treating patients [4][5][6][7]. A recent survey reported by Jereczek-Fossa et al. confirmed that in a highly impacted country like Italy, 73.6% of RT departments shifted to adapted hypofractionated RT schedules [55]. To the best of our knowledge, most guidelines have not focused on PRT, apart from one addressing bone metastases [8]; thus, an RO needing to prescribe PRT during the COVID-19 pandemic would find indications distributed across different papers. The proposed PRT COVID-19 summary (Tables 1 and 2) could aid ROs during this period. The limitation that there are papers missing in our literature search because they were published after our search was conducted exists; however, the average support the summary provides for readers is not compromised and the expert consensus vote poses additional utility, offering an overview perspective for interpreting other similar indications.

P5 Adjuvant (post-surgery) bone metastasis radiotherapy
What is the priority of administering PRT for our patients, both during the COVID-19 pandemic and in the future? Prioritization of PRT has been an object of discussion during the COVID-19 pandemic period [3,9]. Yerramilli et al. suggested prioritizing PRT only for emergencies, providing a triage model to check for the need for PRT [9], although some authors have raised concerns over this [10]. Conversely, Tagliaferri et al. suggested a more inclusive triage-based patient selection strategy, possibly providing PRT even to COVID-19 positive patients, despite the consideration of dealing with highly aggressive diseases such as melanoma [56]. Neither considered the different phases of infection spread within the RT department. Van de Haar et al. [3] suggested multiple phases detailing the steps of expectable crisis from the clinical perspective of RT, surgical, and medical oncology departments, but still did not indicate if or how passing through one of the mentioned level of crisis to another would change the priority list that puts PRT at level four of five in their paper. A few authors [5,6] indicated that different RT and PRT schedules are deliverable in the early or later (more complex) phases (both are included in Tables 1 and 2).
We substantially agreed with the average concern of treating patients during the COVID-19 period. Moreover, most of the indications provided in the early pandemic were unaware of the possible consequent scenarios, thus preparing ahead for the worst possible scenario. A wide range of consequences have been described for RT departments, ranging from compromised [57] to more manageable [58]. In our opinion (when indicated), PRT should remain one of the highest priority treatments from the perspective of ROs. For the two major oncological aims (cure and palliation), the pursued outcomes, as measured by the most appropriate endpoint (i.e., overall survival (OS) for cure and quality of life (QoL) for palliation, respectively), are equivalent from the patient's perspective. Until we are not forced to restrict the delivery of RT to our patients due to the risk of infective spread, palliative and curative settings should be equally prioritized [11]. To put this into context, consider an example on bone-related pain control: if RT is not administered when indicated, the possibly needed dose escalation of medical analgesic therapy can determine side effects affecting QoL, despite controlling pain levels (besides the cost-effective impact on health services by increased drug administration). Separate administration of either palliative RT or medical analgesic therapy should not be considered equivalent by ROs; the concomitant integration of both with modulation over time should be the gold standard.
How to manage the risk of underuse of PRT in the future in patients dealing with complex logistic scenarios? In the future, when the COVID-19 pandemic is over, the issue of patients suitable for PRT who are dealing with complex logistic settings and who are at risk of losing their chance of receiving relief by PRT will surely persist. Looking at the current experience of emergency departments, we are afraid that PRT could be replaced by medical or different alternatives if it is not easily and logistically manageable. Some of the indications provided for PRT during COVID-19 pandemic can be safely and effectively maintained in these peculiar settings. The NORMALITY clinical care model aims to enhance the chance that these patients receive acceptable compromises, aiming for an efficient PRT schedule. The combination of clinical visits, simulation, and RT delivery on the same day is a well-known practice that has diffused over several RT centers for at least 30 years. The Rapid Response Radiotherapy Program (RRRP) was proposed in the literature in the 1990s by the Canadian group Chow et al. [51][52][53]. Similarly, the Vancouver Rapid Access (VARA) for incurable lung cancer was presented by Lefresne et al. [59], as well as the rapid multidisciplinary management of bone metastases described by Donato et al. [60]. The positive impact of the described "advanced practice radiation therapists" on workflows was also explored, for instance, by Job et al. [61,62]. Our model integrates such experiences while focusing on patients with complex logistics, proposing the set of normality model PRT indications for such peculiar settings, as summarized in Tables 3 and 4. If appropriately selected for patients, treatment alternatives such as single fraction treatments applied in emergencies as suggested by Maranzano et al. [20], the use of single repeated schedules as per the "0-7-21" PRT schedule proposed by Nguyen et al. [26], or the bis-in-die (BID) schedules advised by both the "Quad Shot RTOG 8502-QUAD SHOT" report created by Spanos et al. [27,28] and the "Sharon project" for multiple palliative settings [31], can be highly useful. Moreover, our NORMALITY model suggests and offers forms to facilitate the enhancement of preliminary teleconsultations before the  (Figs. 2 and 4). This is in line with the literature acknowledging the efficacy of phone calls [12] and the renewed indication for teleconsultation during the COVID-19 period [9,14]. Some issues remain unaddressed, including the management of patients strictly requiring hospital admittance and the role of technology in balancing urgent palliative patients' needs. Clearly, improving such settings will require multidisciplinary collaboration among operators with different specializations and backgrounds dealing with palliation and oriented to facilitate each other's respective roles and peculiarities.

Conclusion
We provide a comprehensive summary of the literature guideline indications for PRT during the COVID-19 pandemic along with the respective reference and consensus evaluation voted by the AIRO panel. We also propose a clinical care model (based on the clinical guideline indications provided during the COVID-19 pandemic) including clinical indications and written forms facilitating two levels of teleconsultation (triage and remote visits) in order to evaluate patients for indications for PRT ahead of planning live clinical visits. The normality model could facilitate the provision of PRT to patients dealing with future complex logistic scenarios.