Background
Methods
Information sources and search strategy
Study selection
Data extraction and synthesis of results
Statistical considerations | |
1 Basic statistical data provided on incidence of toxicity -Both number of subjects and number of events should be reported -If an estimate of incidence is given the standard error should be supplied | |
2 Numerical labeling of response histogram – if into groups eg. quartiles must state number of patients in each quartile | |
3 When predictive models are correlated with complications parameter estimates must be stated with their standard error | |
4 Complication rates associated with constraints must be reported | |
5 “Goodness of fit” to be reported such as Chi-squared | |
6 Discriminator statistics reported such as receiver operating characteristic curves | |
7 Full organ volumes (rather than partial) should be used -If this is not possible absolute volumes should be used or a standard method of normalization -A clear statement of organ volume definition should be given | |
Toxicity Endpoint considerations | |
1 Symptom-specific information rather than a portmanteau endpoint (eg. RTOG gr 2) should be used | |
2 Consideration that symptoms may be attributed to pre-radiotherapy co-morbidities | |
3 Patient-reporting of symptoms may be important |
Results
Author | Year | Cancer site | No of pts | Pts with tox | OAR studied | Toxicity score used | RT Type | Primary RT Dose (Gy/#) | Pelvic RT dose (Gy/#) | Concurrent chemo use |
Adkison [34] | 2012 | Prostate | 53 | 20 | small bowel | CTCAE v3.0 | IMRT | 70/28 | 56/28 | no |
al-Abany [18] | 2005 | Prostate | 65 | 9 | anal sphincter region | Own questionnaire | 3D | 70.2/39 | NS | no |
Alsadius [19] | 2012 | Prostate or prostatic bed | 403 | 51 | anal sphincter region | Own questionnaire | 3D | 70/35 | NS | no |
Buettner [20] | 2012 | Prostate | 388 | 57 | anal canal | Common grading scheme | 3D | 64/32 or 74/37 | NS | no |
Chopra [11] | 2014 | Cervix (post-op) | 71 | 9 | small bowel, large bowel | CTCAE v3.0 | IG-MRT (46); 3D (25) | 50/25 | 50/25 | 63/71 cisplatin |
Deville [31] | 2010 | Prostate | 30 | 2 | intestinal cavity | RTOG | IMRT | 79.2/44 | 45/25 | no |
Deville [7] | 2012 | Prostatic bed | 36 | 5 | intestinal cavity | RTOG | IMRT | 70.2/39 | 45/25 | no |
Ebert [35] | 2015 | Prostate | 754 | Symptom specific | Anal canal | LENT-SOMA | IMRT | 66–78/33–38 | NS | no |
Fokdal [14] | 2005 | Prostate or bladder | 71 | Symptom specific | small bowel | LENT-SOMA | Conformal | 60/30 (bladder) 69.6/35 (prostate) | 48-60Gy bladder; NS for prostate) | no |
Fonteyne [17] | 2007 | Prostate | 241 | Symptom specific | small bowel, sigmoid | RTOG and “RILIT” | IMRT | 74/37–80/40 | NS | no |
Green [36] | 2015 | Prostate or prostatic bed | 73 | 10 | Intestinal cavity | CTCAE v4.0 | IMRT/VMAT | 61–79.2 | 45 | no |
Guerrero- Urbano [8] | 2010 | Prostate & Pelvic nodes | 79 | 21 | bowel loops | RTOG diarrhoea & LENT SOMA diarrhoea | IMRT | 70/35 | 50/35 or 55/35 | no |
Huang [15] | 2011 | Pancreas | 46 | 8 | duodenum | CTCAE v4.0 | 3D or IMRT | 42/15 | 42/15; 36/15; 38/19 | Gemcitabine; 18 pts. erlotinib in addition |
Isohashi [10] | 2013 | Cervix (post-op) | 97 | 16 | peritoneal cavity, small and large bowel | RTOG/EORTC | 2D or 3D | 50/25 | 50/25 | All nedaplatin |
Kelly [13] | 2013 | Pancreas | 106 | 20 | duodenum | CTCAE v4.0 | 3D or IMRT | 50.4/28 (78pts); 57.5–75.4 in 28–39# (28pts) | 50.4/28 (78pts); 57.5–75.4 in 28–39# (28pts) | Gemcitabine 5-FUor capecitabine +/− cetuximab or erlotinib |
Author | Year | Cancer site | No of pts | Pts with tox | OAR studied | Toxicity score used | RT Type | RT Dose (Gy/#) | Pelvic RT dose (Gy/#) | Concurrent chemo use |
Koper [25] | 2004 | Prostate | 266 | 141 | anal canal | RTOG (simplified) Symptom questionnaire | 3D or 2D | 66/33 | NS | no |
Lind [12] | 2016 | Cervical or Endometrium | 519 | 63 | Anal sphincter, small bowel, sigmoid | Own questionnaire (defecation into clothing without forewarning) | 2D or 3D | 40–46 (endometrium) or 55–70 (cervix) | NS | Not stated |
Mavroidis [27] | 2005 | Prostate | 65 | Symptom specific | anal sphincter | Own questionnaire | 3D | 70.2/39 | NS | no |
Mcdonald [9] | 2015 | Bladder | 47 | 10 | bowel loops | RTOG | 3D | 64/32 | 64/32 | 21 received 5-FU/MMC |
Mouttet-Audouard [6] | 2015 | Cervical | 37 | 8 | Small bowel [defined as peritoneal cavity], sigmoid | CTCAE v4.0 | IMRT (tomotherapy) | 60/28 | 50/28 | Cisplatin |
Peeters [21] | 2006 | Prostate | 641 | 146 | Anal wall | RTOG/EORTC plus 5 specified symptoms | 3D (41 pts. had IMRT boost) | 68/34 or 78/39 | NS | no |
Peeters [28] | 2006 | Prostate | 368 | 32 | Anal wall | Incontinence (no specific questionnaire) | 3D (22 pts. had boost) | 68/34 or 78/39 | NS | no |
Poorvu [16] | 2013 | Cervix or Endometrium (+ PA nodes) | 46 | 3 | peritoneal cavity, small bowel, duodenal segments | CTCAE v4.0 | IMRT | 45/25 (22pts); PAN boost 50–65 (33pts) | 45/25 & PAN boost 50–65 (33 pts) | 24 received cisplatin |
Smeenk [26] | 2012 | Prostate | 48 | 21 | Anal sphincter muscles | Presence of frequency, urgency and incontinence | 3D (n = 43, IMRT (n = 5) | 67.5/27 or 70/28 | NS | no |
Smeenk [22] | 2012 | Prostate | 36 | 23 | Anal wall | Late RILIT score: urgency, incontinence, frequency | 3D | 67.5/27 or 70/28 | NS | no |
Taussky [37] | 2003 | Prostate | 73 | unclear | anal canal | UCLA, FACT-P and EORTC QLQ-PR25 | 3D | 66.6–72/ 37–40 | NS | no |
Thor [29] | 2015 | Prostate | 212 | Symptom specific | Anal sphincter | Own questionnaire with 19 descriptors for 4 symptoms | 3D | 70-78Gy | NS | no |
Verma [30] | 2014 | Cervix & Endometrium | 105 | 9 | duodenum | RTOG and endoscopic findings | IMRT | 45–50 (60-66Gy boost) | 45–50 (60–66 boost) | 58 pts. platinum agents |
Vordermark [23] | 2003 | Prostate or prostatic bed | 44 | 14% severe incontinence | anal canal | 10 question continence questionnaire | 3D | 58–72/29–36 | NS | No |
Yeoh [24] | 2016 | Prostate | 106 | 72% | Anal wall | LENT-SOMA total score | 3D | 66–74.4/ 33–4 | NS | no |
Quality Assessment | |||||||
---|---|---|---|---|---|---|---|
Author | OAR studied | OAR defined | Toxicity definition | Pts with toxicity | Significant findings | Statistical criteria met (1–7) | Endpoint criteria met (1–3) |
Deville [7] | Intestinal cavity | Intestinal cavity below L4–5 | RTOG Gr ≥ 1 | 5/36 (14%) | Toxicity associated with total volume & V20. No constraints specified. | 1,7 (n/a: 2–6) | None |
Mouttet-Audouard [6] | “Small bowel” [outlined as abdominal cavity hence included in this section] | Entire abdominal cavity including all possible organ locations to iliac crests or D12/L1 | CTCAE v4.0 Gr1–3 –diarrhoea or “whole digestive toxicity” (diarrhoea, gastritis, bleeding, pain, incontinence) | 8/37 (21.6%) 17/37 (46%) | Larger volumes of bowel receiving 10–30Gy associated with diarrhoea & whole digestive toxicity. (No constraints specified) “Whole digestive toxicity” associated with many parameters including D20%-D95%. | 1, 7 (n/a 2–6) | 2 |
Green [36] | Intestinal cavity | Not stated | CTCAE v4.0 | 9 (12%) | No dose-volume relationship found. | 1 (n/a 2–6) | 2 |
Deville [31] | Intestinal cavity | Large & small bowel below L4–5 | RTOG Gr ≥ 2 | 2/30 (6%) | No dose-volume relationship found | 1,7 (n/a 2–6) | None |
Isohashi [10] | Peritoneal cavity | Volume surrounding small bowel loops to edge of peritoneum excluding bladder & rectum | RTOG/EORTC Gr ≥ 2 | 16/97 (16.5%) | No dose-volume relationship found | 1,7 (n/a 2–6) | 2 |
Poorvu [16] | 1. Peritoneum 2. Peritoneum + Colon | 1. Possible location of small bowel excluding solid organs & retroperitoneal structures. 2. Peritoneum (as above) plus asc & desc colon | CTCAE v4.0 Gr > 3 | 3/46 (6.5%) | No dose-volume relationship found | 1, 7 (n/a 2–6) | 2 |
Guerrero-Urbano [8] | Bowel loops | Loops from recto-sigmoid junction to 2 cm above PTV | RTOG Gr ≥ 2 diarrhoea; LENT-SOMA consistency & frequency- worst grade | 21/79 (26%) RTOG diarrhoea; ≥gr2 6/79 (7.6%) | V40, V45, V60 and bowel volume of > 450 cc had both higher RTOG & LENTSOMA diarrhoea. Constraints suggested: V40 < 124 cc, V45 < 71 cc, V60 < 0.5 cc for RTOG<gr 2 | 1,7 (n/a 2–3) | 1 |
McDonald [9] | Bowel loops | Loops from recto-sigmoid junction to 2 cm above PTV | RTOG Gr ≥ 1 | 7/47 (14.9%) gr1; 3/47 (6.4%) gr2 | Constraints for < 25% ≥ gr2 toxicity: V30 < 178 cc;V35 < 163 cc;V40 < 151 cc;V45 < 139c; V50 < 127 cc; V55 < 115 cc; V60 < 98 cc V65 < 40 cc | 1,4,7 (n/a 2–3) | 2 |
Chopra [11] | Small bowel | 2 cm above target, individual small bowel loops (unclear how differentiated from large bowel) | CTCAE v3.0 Gr3+ | 9/71 (12.6%) | V15 associated with ≥gr3 toxicity. Recommend V15 < 275 cc, V30 < 190 cc, V40 < 150 cc reduces Gr3 toxicity from 23.6 to 5.6%. | 1, 4, 6 (n/a 2,3) | 2 |
Isohashi [10] | Small bowel | Bowel loops remaining after exclusion of large bowel loops | RTOG/EORTC Gr ≥ 2 | 16/97 (16.5%) | V40 best predictor of late toxicity; Recommend V40 < 340 ml to reduce toxicity from 46.2 to 8.7% | 1,4,6,7 (n/a 2,3) | 2 |
Lind [12] | Small bowel | All visible small bowel in small pelvic cavity to caudal part of sacroiliac joints | Defecation into clothing without warning > 1 in last 6 months | 63/519 (12.1%) | Mean dose>50Gy to small bowel or sigmoid or anal sphincter region associated with symptom (findings for individual organs not clarified) | 1, 7 (n/a 2,3) | 1,2,3 |
Adkison [34] | Small bowel | Not clearly defined | CTCAE v3.0 Gr1 and Gr2 | Gr1 16/53 (30%); Gr2 4/53 (8%) | No dose-volume relationship with V30-V60 small bowel | 1 (n/a 2–6) | None |
Fokdal [14] | Small bowel | Opacified & unopacified small intestine loops (outer contour & contents) from 1st slice to minor pelvis | LENT-SOMA G1–4 | Symptom specific | No dose-volume relationship found | 1,7 (n/a 2–6) | 1,2,3 |
Fonteyne [17] | Small bowel | Not clearly defined | RTOG and “RILIT” Gr1 & Gr2 | Gr1 112/241 (46%), Gr2 32/241(13%) | No dose-volume relationship found | 1, (n/a 2–6) | 1,2 |
Poorvu [16] | Small bowel | Opacified & non-opacified small bowel loops | CTCAE v4.0 Gr3+ | 3/46 (6.5%) | No dose-volume relationship found | 1,7 (n/a 2–6) | 2 |
Huang [15] | Duodenum | Duodenal bulb to ligament of Treitz | CTCAE v4.0 Gr ≥ 3 | 8/46 (17.4%) | With a V25 > 45% toxicity rates increase from 8 to 48% | 1,4,6,7 (n/a 2,3) | 2 |
Kelly [13] | Duodenum | Gastric pylorus until end of duodenum 3 cm past midline | CTCAE v4.0 Gr ≥ 2 | 20/106 (18.9%) | With a V55 > 1 cc toxicity rates increase from 9 to 47% | 1,4,6,7 (n/a 2,3) | 2 |
Verma [30] | Duodenum | From gastric outlet through transverse portion of duodenum (ascending portion excluded) | RTOG, all grades | 9 /105 (8.6%) | With a V55 > 15 cc toxicity rates increase from 7.4 to 48.6% | 1,4,6,7 (n/a 2,3) | 2 |
Poorvu [16] | Duodenal segments | D1 segment: bulblike shape & origin beyond gastric pylorus. Transitions between 2nd & 3rd segments was lateral border of IVC; Between 3rd & 4th was medial border of aorta | CTCAE v4.0 Gr ≥ 3 | 3/46 (6.5%) | No dose-volume relationship found with duodenum | 1,7 (n/a 2–6) | 2 |
Quality Assessment | |||||||
---|---|---|---|---|---|---|---|
Author | OAR studied | OAR defined | Toxicity definition | Pts with toxicity | Significant findings | Statistical considerations met (1–7) | Endpoint considerations met (1–3) |
Chopra [11] | Large bowel | 2 cm above target, individual loops of large bowel (unclear how differentiated from small bowel) | CTCAE v3.0 Gr ≥ 3 | 9/71 (12.6%) | V15 associated with ≥gr 3 toxicity. Constraints: V15 < 250 cc, V30 < 100 cc, V40 < 90 cc to reduce toxicity from 26.7 to 5.4% | 1, 4, 6 (n/a 2,3) | 2 |
Isohashi [10] | Large bowel | Single loop continuing from end of sigmoid to ascending colon | RTOG/EORTC, Gr ≥ 2 | 16/97 (16.5%) | No constraint found for large bowel | 1,7 (n/a 2–6) | 2 |
Fonteyne [17] | Sigmoid colon | Where rectum sweeps anteriorly to one slice above aortic bifurcation | RTOG and “RILIT” Gr 1 and 2 | Gr 1112/241 (46%), Gr 2 32/241 (13%). | V40 associated with gr1 diarrhoea & blood loss. Constraints: V40 < 10%, V30 < 16% to avoid gr1–2 diarrhoea | 1, 7 (n/a 3) | 1,2 |
Mouttet-Audouard [6] | Sigmoid colon | Anterior curvature of sigmoid colon to anterior abdominal wall | CTCAE v4.0 Gr1–3 diarrhoea and “whole digestive toxicity” | 8/37 (21.6%) diarrhoea; 17/37 (46%) (whole tox) | ‘Whole late digestive toxicity’ associated with V30–40. No specific constraints. | 1,7 (n/a 2–6) | 1,2 |
Lind [12] | Sigmoid colon | From where rectum deviates from its mid- position to where it turns cranially in left abdomen connecting to colon descendens | Defecation into clothing without warning > 1 in last 6 months | 63/519 (12.1%) | Mean dose>50Gy to small bowel or sigmoid or anal sphincter region associated with symptom (findings for individual organs not clarified) | 1, 7 (n/a 2–6) | 1,2,3 |
al-Abany [18] | Anal sphincter region | Caudal 3 cm of the rectum from anal verge (including filling) | Own questionnaire; Faecal leakage >2X/week | 9/65 (13.8%) faecal leakage | Increased risk with mean dose of 45-55Gy. Constraints: V35 < 60%, V40 < 40% associated with no risk of faecal leakage. | 1, 7 (n/a 2,3) | 1,2,3 |
Alsadius [19] | Anal sphincter region | Caudal part of large bowel, from end of rectal ampulla where bowel no longer had visible content or air. | Own questionnaire; Faecal leakage >once per month | 51/403 (12.7%) faecal leakage | Dmean<40Gy reduces risk from 17 to 4%. | 1,2,4,7 (n/a 3) | 1,2,3 |
Fokdal [14] | Anal canal | Outer contour of the structure extending from anal verge 2 cm cranially | LENT SOMA score | Urge: 27/71 (38%); Incontinence: 21/71 (30%) | Urgency related to Dmed> 33.8: increases toxicity 31 to 47% Incontinence related to Dmax> 53.8 increases 14 to 44% | 1,2,4,5,7 (n/a 3) | 1,2,3 |
Vordermark [23] | Anal canal | Anal verge to the section below visible rectal lumen, corresponding to the upper border of the levator ani muscle | “Solid soiling” (Severe incontinence) Own continence questionnaire | 6/44 (14%) | Severe incontinence - associated with Dmin (23.1Gy) - related to portals extending 2 mm below ischial tuberosities (compared with 5 mm above) | 1, 7 (n/a 2–3) | 1,2,3 |
Koper [25] | Anal canal | Caudal 3 cm of the intestine | RTOG gr1 + 2; Plus symptom questionnaire. | 141/248 (57%) | D90% (=54.9Gy) to associated with ≥ gr1 rectal toxicity | 1, 7 (n/a 2–6) | 2,3 |
Taussky [37] | Anal canal | Most distal 2-3 cm of rectum | 10 questions from UCLA-PCI, FACT-P & EORTC QLQ -PR25 | Unclear | no relation with anal canal DVH found | 7 (N/a: 2–3) | 3 |
Buettner [20] | Anal canal | Caudal 3 cm of rectum | Common grading scheme; subjective sphincter control at highest grade | 57/388 (14.7%) | DSH data: Toxicity correlated with dose to anal surface: lateral extent 53Gy > 56%. DVH data: Dmean 47Gy to anal sphincter volume correlated with sphincter toxicity. Constraints: Dmean<30Gy.NTCP modeling to LKB model TD50 = 120, m = 0.42. | 1,3,6,7 (n/a 2) | 1,2,3 |
Peeters [21] | Anal wall | Wall of caudal 3 cm of anorectum (method described) | RTOG/EORTC ≥ gr 2 and ≥ gr 3 Plus incontinence pad use>2x/wk. | ≥gr 2165/641 (25.7%) ≥ gr 3 27/641 4.2% | Dmean increase from 19Gy to 52Gy increased gr2 toxicity: 16 to 31%. V65 & Dmean most significant for incontinence. Dmean increase by 33Gy increased incontinence by 12% | 1,2,4,6,7 (n/a 3) | 1,2 |
Mavroidis [27] | Anal sphincter region | Musculaure layer around the rectal aperture, 3 cm caudal from anal verge | Own questionnaire | faecal leakage 19/65 (29%); blood/mucus 22/65 (34%) | Relative seriality NTCP model of anal sphincter for incontinence, blood/mucus. Parameters for incontinence: D50 = 70.2Gy, γ = 1.22, s = 0.35. Parameters for blood/mucus: D50 = 74.0Gy, γ = 0.75, s ≈ 0 | 1, 3, 5, 6, 7 (n/a 2) | 1,3 |
Peeters [28] | Anal canal wall | Wall of caudal 3 cm of anorectum (method described) | Incontinence requiring pad use>2x/wk.; | 32/368 (7%) | NTCP LKB model of incontinence with anal wall dose. Parameters found were n = 7.48; TD50 = 105; m = 0.46 | 1,3,4,5,6,7 (n/a: 2) | 1,3 |
Smeenk [26] | Anal sphincter muscles | Individual muscles defined (Internal anal sphincter (IAS), external anal sphincter (EAS), puborectalis & levator ani) | Frequency, Urgency, Incontinence | 21/48 (44%) | For complication <5% Dmean<30Gy to IAS; <10Gy to EAS, < 50Gy to puborectalis, <40Gy to levator ani | 1, 4,5 (n/a 2) | 1,2,3 |
Smeenk [22] | Anal wall | Continuation of rectal wall from anal verge to slice below lowest slice with a rectal balloon | Frequency, urgency, incontinence | 39% frequency, 31% urgency, 31% incontinence | For urgency: Anal wall Dmean<38Gy risk < 15%, >38Gy risk is 62% | 1,4,7 (n/a 2,3,5,6) | 1,3 |
Lind [12] | Anal sphincter region | Inner muscle layer of the sphincter up to anal verge | Defecation into clothing without warning > 1 in last 6 months | 63/519 (12.1%) | Mean dose>50Gy to small bowel or sigmoid or anal sphincter region associated with this symptom (findings for individual organs not clarified) | 1, 7 (n/a 2,3) | 1,2,3 |
Yeoh [24] | Anal wall | From anorectal junction (not clearly defined) | LENT-SOMA total score | 72% | Anal wall V40 > 65% associated with chronic toxicity. | 1,5 (n/a 2,3) | 2,3 |
Thor [29] | Anal sphincter | Anal canal, inner and outer sphincter (not clearly defined) | Questionnaire of 19 questions in 4 domains: pain urgency, mucus & incontinence. | Specific to each of 19 question | 5 LKB models proposed for anal sphincter doses. Low anal sphincter dose associated with faecal leakage and pain. High anal sphincter dose associated with leakage. | 1,3,6 (n/a 2) | 1,2,3 |
Ebert [35] | Anal Canal | Caudal 3 cm of anorectum | LENT-SOMA – 8 symptoms | Specific to each symptom | Bleeding associated with >40Gy, proctitis with 36-63Gy, frequency with 8-85Gy, urgency and tenesmus with 5-34Gy to anal canal. | 1,5,7 (n/a 2) | 1,2,3 |
Whole bowel
Peritoneal cavity
Bowel loops
Small bowel and its components
Small bowel
Duodenum
Large bowel and its components
Large bowel and sigmoid Colon
Anal canal
Dmean
Study | No of pts | OAR | Endpoint | Dmean (in EQD2) constraint | Risk of endpoint below this constraint | Risk of endpoint above this constraint |
---|---|---|---|---|---|---|
Al-albany [18] | 65 | Anal sphincter region | Incontinence >2X/week | 43.2 | 8% | 52% |
Alsadius [19] | 403 | Anal canal | Incontinence > 1x/month | 40 | 5.2% | 21% |
Buettner [20] | 388 | Anal sphincter region | Incontinence: moderate/severe (gr2) | 47, though <30Gy ideal | 5% (approx; read from graph) | |
Smeenk [22] | 36 | Anal canal wall | Urgency present | 41.8 | 15% | 62% |
Peeters [21] | 641 | Anal canal wall | Incontinence requiring pad >2x/week | No constraint specified | 16% at 19Gy | 31% at 52Gy |