COMBINATION OF DWI/DCE-MRI AND [18F]FECH PET/CT IN PATIENTS WITH BIOCHEMICAL RELAPSE OF PROSTATE CANCER: IMPACT ON TREATMENT DECISION MAKING

1

INDEX

Summary List of Abbreviations

Introduction The Biochemical Recurrence

Aim Of The Study Study Design Patients And Method

Results Discussion Conclusion References pag. 2 pag. 5 pag. 7 pag. 18 pag. 27 pag. 29 pag. 32 pag. 40 pag.56 pag.61 pag.64

2

SUMMARY

Introduction: Worldwide prostate cancer is the most commonly diagnosed cancer in men, remaining the second most common cause of cancer death in men in the developed world. In recent years more and more of prostate cancer patients from the western hemisphere are diagnosed at an earlier stage due to rising prevalence of prostate-specific antigen (PSA), being consequently treated with radical intent (either surhery or radiotherapy). Despite the technical improvement in screening and therapies, between 27-53% of all patients underwent radical prostatectomy or radiation therapy will develop PSA-recurrence, and second-line treatment will then be required in 16%-35% of cases. In these patients, distinguishing between local and distant failure is crucial to choose the most suitable treatment.

Aim: We aimed to investigate the impact of DWI/DCE-MRI and [18F]FECH PET/CT in treatment-decision making in patients with prostate cancer recurrence after a curative first-line treatment, demonstrating that an accurate pre-treatment restaging using the combination of these new functional imaging modalities may be used to optimize treatment decision making in patients with recurrent PCa, improving final outcome. Imaging techniques were used to identify sites of relapse, define different pattern of recurrence (loco-regional or distant metastasis), determining the rate of patients presenting recurrence in form of local relapse (suitable for lodge

3

radiotherapy), oligometastatic disease (suitable for Hazard Ratio Image-Guided Radiation Therapy, HR IGRT), or plurimetastatic disease (candidate for androgen deprivation therapy, ADT).

Methods:We enrolled 30 patients with biochemical relapse of prostate cancer who underwent two consecutive imaging procedures: f-MRI and [18F]FECH PET/CT, for a total of 60 exams. Each exam was classified as follow: negative/positive/indeterminate, and results were used to classify patients in the following groups: patients with local relapse, patients with oligo-metastatic recurrence(≤5 total lesions detected) and patients with oligo-metastatic disease(≥5 total lesions detected). Accordingly, the classification derived from the combination of both f-MRI and [18F] FECH PET/CT guided the selection of the most suitable as previously explained.

Results: Combining the informations derived by the two methods, we were able to classified patients as: patients with exclusive local relapse (n=7), who underwent salvage EBRT therapy; patients with oligometastatic disease (≤5 total lesions, n=18) who underwent respectively: pelvic irradiation (n=4), pelvic + nodal irradiation with HR IGRT (patient with oligometastatic loco-regional disease, n=4), HR IGRT on bone (n=3), a combination of ADT and IGRT on bone (n=3) and a combination of bone HR IGRT+ lodge RT (patients with oligometastatic diseases involving bone, n=4); patients with plurimetastatic disease underwent ADT (n=2).

4

Even though respectively 33% patients resulted negative by [18F]CH PET/CT

scan and 27% patients by DWI/DCE MRI scan,

matching the total findings only 10% patients resulted definitively negative. This dual-approach changed patients findings in 80% totally, with a separate contribute of 43% patients with more lesions detected by [18F]-FECH PET/CT and 37% patients with more lesions detected by DWI/DCE MRI. Conclusions: The follow-up data showed a 74% of total positive responses (included both recovery and decreased PSA value), confirming that the high detection rate reached by combining DWI/DCE-MRI and [18F]FECH PET/CT impacted positively on patients' management, by increasing the number of patients treated with EBRT/HD IGRT instead of ADT (expensive and with lots of side effects) and by a more accurate definition of the number of lesions included in the radiation planning.

5

LIST OF ABBREVIATIONS

Apparent Diffusion Coefficient Androgen Deprivation Therapy Computed Tomography

Dynamic-Contrast Enhanced Disease Free Survival

Digital Rectal Examination Diffusion-Weighted Imaging European Association of Urology External Beam Radiotherapy Fluorodeoxyglucose

Fluoroethylcholine

Functional Magnetic Resonance Imaging Gleason Score

High Dose

Hazard Ratio Image-Guided Radiation Therapy Intensity Modulated Radiotherapy

Rate Constant Transfer Constant

Lower Urinary Tract Symptoms Magnetic Resonance Imaging Overall Survival

Prostate Cancer

Prostate Specific Antigen Positron Emission Tomography

6 RP RT SBRT SUV TRUS TURP ve VIMAT Radical Prostatectomy Radiation Treatment

Stereotactic Body Radiotherapy Standard Uptake Value

Transrectal Ultrasonography

Transurethral Resection of the Prostate Fractional Volume

7

INTRODUCTION

PROSTATE CANCER

Malignant neoplasms of the prostate, referred as prostate cancer (PCa, ICD-10 C61) usually originate in the glandular tissue (adenocarcinomas) and are most often located in the peripheral zone of the prostate.

Epidemiology: worldwide PCa represents the most common cancer in men

and the second cause of cancer death in men in the developed world1, exceeded only by lung cancer. European trends confirm these data: in 2008 about 371,000 Europeans were newly diagnosed with PCa, accounting for 22% of all cancer diagnoses in male2, with an actual incidence rate varying from 25 to 193/100,000 men-years3. In almost all European countries, PCa mortality rose in the 1980s and the first part of the 1990s4.

The estimated age-standardized incidence rate for PCa in Italy is 108.4/10000 representing 11% of all cancers in male5. The mortality trends slightly increased from 1970’s to 1990’s and then started to decrease, being expected down to 14 per 100,000 person-years in 2015, with an estimated reduction of about 25% with respect to the peak level6,7 (Fig.1). Moreover, last data on survival reported a percentage of 5-years survival ranging from 80-88% in the period of 1995-20048,9, even more of the survival of the majority of European countries (70-80%)10,11.

8

Figure 1 Trends on Incidence/Mortality of PCa in Italy; data are considered on 100.000 people (adapted from Registro Tumori 2012)

Risk Factors: For a disease with such high prevalence and incidence as PCa

is, relatively little is known about its exact etiology. While the etiology and pathogenesis of PCa are clearly complex and multifactorial, the variable risk in individuals and populations has revealed a number of important risk factors. However, convincing evidence has been produced for only few risk factors, such as age12-14_{, family history and genetics}15-18_{, ethnicity,}

androgenetic status19, diet20-22, chronic inflammation/infection23,24 (Fig. 2) and anthropometric factors such as obesity25-30.

Other factors considered of importance in PCa development are the use of other hormones such as leptin30-33, or some habits such as cigarette smoking34,35. 0 20 40 60 80 100 120 140 1980 1985 1990 1995 2000 2005 2010

Prostate Cancer in Italy

9

Figure 2 Multi-step process of prostate cancer development. Under normal conditions, Th cytokines are

maintained in a homeostatic state via self-regulating mechanisms and angiostatic chemokines are predominant. Regions of PIA are frequently associated with inflammation, possibly triggered by an infectious agent. The inflammatory response induces TLR-expressing inflammatory cells, which mediate cell

proliferation and increase cytokine and chemokine production. As the inflammatory response progresses, self-regulating mechanisms fail leading to an overproduction of Treg, Th17 cytokines and angiogenic chemokines, which enhance DNA damage, cell proliferation and angiogenesis, promoting prostate cancer

progression (from Tindall EA, Cancers 2010 2)

Clinical Presentation: From a clinical point of view, PCa presentation might

be with disease at any stage: very frequently does not cause any symptoms at all36 but when urethral compression is present as a consequence of enlarge prostate gland, patients will present with symptoms related to obstruction. These symptoms are generally described as LUTS (lower urinary tract symptoms).

Diagnosis and Staging: In recent years, more and more patients with PCa in

the western hemisphere are diagnosed due to rising prevalence of prostate-specific antigen (PSA) testing37. Even if the use of PSA as screening test has the potential drawback of overdiagnosing the disease (i.e. in patients with indolent tumors that would die for other causes)38,39 most patients are

10

diagnosed at an earlier stage. Today, the diagnosis of PCa results more on the realization of digital rectal examination (DRE) and PSA blood test leading to randomized transrectal ultrasonography (TRUS) biopsy than on other clinical symptoms40. Usually an elevated level of PSA evokes a diagnosis of PCa, especially if the DRE is suspect. The level of PSA is a continuous parameter: the higher the value, the most likely the existence of PCa. PSA levels, anyway, are not correlated with the tumoral aggressiveness, currently identified by a Gleason score >7 on the biopsies41. Gleason Score is the most commonly used pathologic grading system, first described by Donald F. Gleason in 1966 42,43 (Table 1 and Fig.3).

Table 1 The original Gleason Score classification according to Gleason (1966)

THE 5 PATTERNS ACCORDING TO THE ORIGINAL GLEASON SCORE

1 Very well differentiated, small, closely-packed, uniform, glands in essentially

circumscribed masses

2 Similar (to pattern 1) but with moderate variation in size and shape of glands and _{more atypia in the individual cells; cribriform pattern may be present, still}

essentially circumscribed, but loosely arranged

3

Similar to pattern 2 but marked irregularity in size and shape of glands, with tiny glands or individual cells invading stroma away from circumscribed masses, or

solid cords and masses with easilyidentifiable glandular differentiation within most ofthem

4 Large clear cells growing in a diffuse pattern resembling hypernephroma; may

show gland formation

5 Very poorly differentiated tumours; usually solid masses or diffuse growth with

11

In 2005 ISUP modified Gleason System44-46 was introduced with a main aim of clarifing some controversial aspects of Gleason original classification, especially regarding pattern 3 and 4 (Table 2).

Table 2 The 5 bioptic patterns according to Gleason score modified system ISUP 2005. Adapted from

Montironi, Diagnostic Histopathology 2011(43)

Figure 3 Schematic histological difference between 5 Gleason score patterns

THE 5 PATTERNS ACCORDING TO THE 2005 ISUP MODIFIED SYSTEM

1 Circumscribed nodule of closely-packed but separate, uniform, rounded to

oval, medium-sized acini (larger glands than pattern 3)

2

Like pattern 1, fairly circumscribed, yet at the edge of the tumour nodule there may be minimal infiltration. Glands are more loosely arranged and not quite as uniform as Gleason pattern 1

3

Discrete glandular units. Typically smaller glands than seen in Gleason pattern 1 or 2. Infiltrates in and amongst nonneoplastic prostate acini. Marked variation in size and shape. Smoothly circumscribed small cribriform nodules of tumour

4

Fused microacinar glands. Ill-defined glands with poorly formed glandular lumina. Large cribriform glands. Cribriform glands with an irregular border. Hypernephromatoid

5

Essentially no glandular differentiation, composed of solid sheets, cords, or single cells. Comedocarcinoma with central necrosis surrounded by papillary, cribriform, or solid masses

12

Apart from guiding biopsies, imaging plays a fundamental role for tumor staging. The TNM staging system 200947 is the most commonly used method for PCa staging classification. These staging criteria are designed to serve several purposes: helping to predict a patient’s prognosis, assisting in the planning of treatment strategies, and providing a common language for practitioners to report the extent of disease (Table 3). Prognostic groups are further stratified based on TNM, according to the d’Amico classification48or the recent European Association of Urology (EAU) guidelines, further descrive four prognostic groups49.

The primary tumor (T) is initially assessed by a combination of DRE and TRUS50,51 with or without Color Doppler US and Power Doppler52. Magnetic Resonance Imaging (MRI) is currently the most accurate noninvasive method for the detection and staging of localized PCa53-56.

Regarding lymph node status (N), both MRI and CT have a limited role due to their intrinsic limitation of considering as pathological any enlarged lymph-nodes with size > 1 cm. In this field, PET/CT with radiolabelled choline, a radiopharmaceutical that accumulates into the membrane of active proliferating tumor cells, has shown promising results for pre-operative lymph node staging57 however, sensitivity of this technique is stristcly dependent of initial PSA levels58,59.

13

Table 3 TNM staging system according to the ultimate revision of 2009 (from Sobin TNM Classification of

Malignant Tumours, 7th Edition 2009)

The “M” parameter, representing metastatic spread of PCa is generally performed by bone scintigraphy60. In fact, the main site of distant metastasis in PCa is bone, with bone involvement present in 90 % of patients dying for

TNM STAGING SYSTEM

T

TX: cannot evaluate the primary tumor imaging T0: no evidence of tumor

T1: tumor present but not detectable clinically or with imaging

T1a: tumor incidentally found in < 5 % of prostate tissue resected (for other reasons)

T1b: tumor was incidentally found in greater than 5 % of prostate tissue resecate

T1c: tumor was found in a needle biopsy performed due to an elevated serum PSA

T2: the tumor can be felt on examination but has not spread outside the prostate

T2a: the tumor is in half or less than half of 1 of the prostate gland’s 2 lobes

T2b: the tumor is in more than half of one lobe but not both

T2c: the tumor is in both lobes

T3: the tumor has spread through the prostatic capsule

T3a: the tumor has spread through the capsule on one or both sides

T3b: the tumor has invaded one or both seminal vesicles

T4: the tumor has invaded other nearby structures (rectum, pelvic side wall)

N

NX: cannot evaluate the regional lymph nodes

N0: there has been no spread to the regional lymph nodes N1: there has been spread to the regional lymph nodes

M

MX: cannot evaluate the distant metastases M0: no distant metastases

M1: distant metastasis

M1a: the cancer has spread to lymph nodes beyond the regional ones M1b: the cancer has spread to bone

14

PCa61,62. Bone scan sensitivities are higher as compared to PSA and plain radiographs63,64, however the main benefit of the technique is in the initial assessment of high-risk patients presenting with PSA > 20 ng/ml, Gleason score > 8, bone pain or stage T3/T4 disease65.

PET/CT with radiolabelled choline has been used for “M” staging in high risk patients, resulting in a 20% change of patients’ therapeutic management66.

Therapies: Once the diagnosis is reached, patients with no advanced disease

are treated with either surgery67-71 or radiation therapy72-75 (RT,as external beam or transperineal brachytherapy76-78), alone or in combination79. Androgen Deprivation Therapy (ADT)80-82 is then generally performed. Active surveillance to monitor disease progression and treat only when tests cancer progression is deserve to patients with low risk83.Table 4 is a summary

15

STAGE T1a

Watchful Waiting In patients with <10-year life expectancy standard treatment for Gleason score ≤6 and 7 adenocarcinomas

Active Surveillance In patients with >10 years life expectancy, re-staging wuth TRUS and biopsy is recommended

Radical Prostatectomy

Optional in younger patients with a long life expectancy, in particolar in poorly differentiated tumours. Higher complications risks after TURP, especially with interstitial radiation

Radiotherapy

Optional in younger patients with a long life expectancy, in particolar in poorly differentiated tumours. Higher complications risks after TURP, especially with interstitial radiation

Hormonal Not an option

Combination Not an option

T1b-T2b

Watchful Waiting Patients with life expectancy <10 years Active Surveillance

Treatment options in patients with cT1c-cT2a, PSA <10 ng/ml, GS ≤6, ≤2 biopsies positive, ≤ 50 %cancer involvement of each biopsy

T1a-T2c

Watchful Waiting Patients with life expectancy <10 years and GS ≤7 Radical Prostatectomy

Optional in patients with pT1aPCa. Standard treatment for patients with a life expectancy > 10 years who accepted treatment-related complications

Radiotherapy

Patients with life expectancy >10 years who accepted treatment-related complications

Patients with contraindications for surgery

Unfit patients with 5-10 years of life expectancy and poorly differentiated tumours (combination therapy is recommended)

Brachiterapy

Low-dose rate brachytherapy can be considered for low risk PCa patients with a prostate volume ≤50 mL and an IPSS≤12

Hormonal

Symptomatic patients, who need palliation of symptoms, unfit for curative treatment

Anti-androgens are associated with a poorer outcome compared to “watchful waiting” and are not recommended Combination For high-risk patients, neodjuvant ADT and concomitant

16

T3-T4

Watchful Waiting

Option in asymptomatic patients with T3, Gleason score ≤7, and a life expectancy <10 years who are unfit for local treatment

Radical Prostatectomy

Optional for selected patients with T3a, PSA <20 ng/ml, GS ≤8 and a life expectancy > 10 years

Patients have to be informed that RP is associated with an increate risk of positive surgical margins, unfavourable hystology and positive lymph nodes and that, adjuvant or salvage therapy such ase RT or ADT might be indicated Radiotherapy

T3 with >5-10 years of life expectancy. Dose escalation of >74 Gy seems to be of benefit. A combination with ADT can be recommended

Hormonal

Symptomatic patients, extensive T3-T4, high PSA level (>25-50 mL), PSA DT <1 yr

Patients-driven, unfit patients

Hormone monotherapy not an option for patients who are fit enough for radiotherapy

Combination

Overall survivor is improved by concomitant and adjuvant ADT (3 years) combined with external beam radiation NHT plus RP: no indication

N+, M0

Watchful Waiting Asymptomatic patients. Patient-driven (PSA<20-50 ng/ml), PSA DT >12 months. Requires very close f- up

Radical Prostatectomy Optional for higly selected patients with a life expectancy of >10 years as part of a multimodal treatment approach

Radiotherapy

Optional in higly selected patients with a life expectancy of >10 years. Combination therapy with adjuvant ADT for 3 years is mandatory

Hormonal

Standard treatment after estended node dissection if >2 positive nodes (irrespective of the local treatment: surgery or RT). It should be used only as mootherapy in patients who are unfit for any type of local therapy

Combination No standard option. Patient-driven M+

Watchful Waiting

No standard option. May have worse survival/more complications than with immediate hormonal therapy. Requires very close f-up

Radical Prostatectomy Not a standard option Radiotherapy

Not an option for curative intent; therapeutic option in combination with androgen deprivation for treatment of local cancer-derived symptoms

Hormonal Standard option. Mandatory in symptomatic patients

17

Other alternative local treatment options have emerged, such as cryosurgical ablation84 or HIFU85, but their role still need to be clarified.

Follow-up strategies: Patients treated for PCa are usually followed lifelong

by serum PSA determinations, disease specific history, and digital rectal examination. Specific follow-up intervals and the number of tests have not been jet established often are individualized86 (as shown in Table 5). Patients initially managed by active survillance need individual follow-up based on treatment intent and tumor characteristics. Routine use of imaging procedures is not recommended and should only be limited to specific situations.

The controversies related to follow-up were partially solved by EAU guideline 201486 (Table 5).

Therefore, clinicians are often guided by their own experience, also depending on the primary treatment used. Follow-up must be accurate as best as possible because the significant risk of cancer recurrence after therapy remains high87-89.

18 1

In asymptomatic patients, a disease-specific hystory and a serum PSA measurement supplemented by digitl rectal exams are the recommended tests for routine follow-up. These should be performer at 3,6,12 months after treatment, then every 6 mo until 3 ys, and then annually

2 After radical prostatectomy, a serum PSA level >0.2 ng/ml can be associated with

residual or recurrent disease

3

After radiation therapy, a rising PSA level 2.0 ng/ml above the nadir value, rather than a specific thershold value, is the most reliable sign of persistent or recurrent disease

4 Both a palpable nodule and a rising serum PSA level can be signs of local disease

recurrence

5

Detection of local recurrence by TRUS and biopsy is only recommended if it will affect the treatment plan (salvage,radiotherapy, or surgery). In most cases TRUS and biopsy are not necessary before second-line therapy

6

Metastasis may be detected by pelvic CT/MRI or bone scan. In asymptomatic patients these examinations may be omitted if the serum PSA level is <20 ng/ml, but data on this topic are sparse

7

Routine bone scans and other imaging studies are not recommended in asymptomatic patients. If a patients has a bone pain, a bone scan should be considered irrespective of the serum PSA level

Table 5 EAU 2014 recommendation for follow-up in patients with PCa after radical treatment

THE BIOCHEMICAL RECURRENCE

Following radical prostatectomy (RP), biochemical relapse is defined as a confirmed PSA value >0.2 ng/ml by two consecutive increases90. Following RT, a PSA value of 2 ng/ml above the nadir after RT represents recurrent cancer91.

About 27-53% of all patients treated with RP or RT will develop PSA-recurrence, thus requiring second-line treatment in 16%-35% of cases92,93. The rate of biochemical relapse after RP or RT is particularly high in patients with stage T3 disease and positive resection margins at time of initial treatment94,95.

19

When relapse occurs, the definition of isolated local failure versus distant failure is critical. Initial pathology, the time after primary therapy when PSA-relapse occurs, the velocity of PSA rising (PSA doubling time) are useful parameters in the distinction between local and distant failure. Indeed, poorly differentiated tumour, early PSA relapse, and a quickly rising PSA are most likely to be signs of distant failure (systemic disease), whereas in patients with moderately differentiated tumours, late PSA relapse, and a slow doubling time (>10–12 mo) local failure only can be hypothesized96.

Such distinction is significant since treatment is established according to the site of disease relapse, the patient’s general condition and patients’ personal preferences. Therefore, accurate restaging is mandatory for a correct treatment planning.

TRUS is neither sensitive nor specific in detecting local recurrence after RP or RT97. Similarly, sensitivity of anastomotic biopsies remains low98.

Imaging such as bone scintigraphy or CT are of poor diagnostic value unless the PSA serum level is > 20 ng/ml or unless the PSA velocity is > 2 ng/ml/yr99,100.

MRI with endorectal coil imaging represents a useful technique to detect local recurrences if the PSA serum level is > 2 ng/ml101. Several studies have reported promising results in detecting local recurrence using MRI, with sensitivities and specificities of 84-88% and 89-100% respectively102,103. However, MRI accuracy for extra-local disease is still to be determined104.

20

PET with radiolabelled choline has been successfully applied for detecting the sites of PCa relapse105, but contradictory results have been reported at low PSA levels59,106-108. Moreover, the physiological urinary excretion of radiolabelled choline (higher for fluorocholine derivatives as compared to [11C]-Choline) can limit the diagnostic accuracy when post-surgical residual tissue is present. In case of nodal involvement110 and metastatic disease111, on the contrary, PET/CT with radiolabelled choline was found to have a role in disease detection, that is fundamental to define the most suitable therapy (local salvage therapies versus systemic therapies112). The detection rate of PET/CT with radiolabeled choline in patients with biochemical recurrence shows a linear correlation with PSA value and reaches about 75% in patients with PSA > 3 ng/ml113. Sensitivity, specificity and accuracy for detecting body lymph-nodes or bone metastases are 92.3%, 100% and 93.8%, respectively114.

On the basis of these considerations, it was reported that up to 23% patients with initial PSA relapse115 (with low PSA levels) might remain with unknown sites of recurrence at the conventional imaging techniques116,117.

This finding explains why none of the main International Guidelines at present recommends any imaging procedure if:

a) PSA does not reach high levels;

b) there is no evidence of a clinically detectable lesion (pain, fracture,etc) that can be treated with palliative intent.

21

Therefore, the selection of the imaging procedure is decided on patient-basis. In addition to this problem selection of the timing and treatment modality of PSA-only recurrence remain controversial and is itself influenced by the possibility to identify relapse sites.

When signs of local recurrence are present (i.e. low-risk patients with late recurrence and a slow PSA rise) patients with PSA failure who are not opting for second-line curative options are best managed by observation alone115. Isolated local relapse potentially confined to the pelvis can be cured by salvage external beam radiotherapy (EBRT) while salvage RT is deserved to patients with organ-confined disease, with a low co-morbidity and a life expectancy of at least 10 years117. The decision to undergo a salvage treatment must be evidence-based based on several parameters such as the patient’s expected survival, the natural history of his recurrence, the presence of local or a systemic relapse and treatment goal (overall survival benefit, symptom benefit, symptom-free duration benefit, side-effects)118.

When multiple distant metastasis are suspected, local therapy is not performe (except for palliative reasons) and despite controversial results, early ADT is generally performed119. In fact, in presence of systemic relapse early ADT represents the best treatment, affecting progression of disease in high risk patients (i.e. Gleason Score > 8). Some retrospective studies have demonstrated that there is not a significant difference in the time to clinical metastases with early ADT (after PSA recurrence, but before clinical

22

metastases) versus delayed ADT (at the time of clinical metastases)119. However, no overall impact of early ADT on PCa-specific mortality has been demonstrated120.

Intermittent androgen deprivation (IAD) may be an alternative to delay the time to androgen independence and hormone-refractory disease, to minimize side-effects and to reduce prolonged therapy costs121.

ASCO guidelines122 concluded that no recommendations can be made on when to start hormonal therapy in asymptomatic patients until data becomes available from studies using modern diagnostic and biochemical tests and standardised follow-up schedules.

Chemotherapy is an option for patients with hormone-refractory PCa123, but when to initiate a cytotoxic regime remains controversial124.

Rethinking the treatment strategies of PCa recurrence has been crucial in these last years, for the need of coupling site specific therapeutic approaches with the possibility of better localize disease relapse111.

A new concept of disease relapse has been introduced, namely the definition of “oligometastic disease” when < 5 lesions in a patient are detected125

. In this circumnstance, patients are susceptible to treatment stereotactic body radiotherapy with local control rates of about 80%126. In fact, due to the delivery of higher radiation doses to the tumour without normal tissues toxicity, intensity modulated RT (IMRT) and high-dose volumetric arc IMRT with image-guided RT (HD IGRT) has transformed the concept of “salvage

23

radiotherapy” to “curative” radiotherapy127

. Three dimensional treatment planning allows for the design of complex treatment plans that conform closely to the high-dose volume and to the shape of the tumour128-131.

To differentiate patients candidate for local salvage treatment and those who might need systemic therapy a comprehensive diagnostic work-up should be guarantied;this would imply that we need technologies to localize disease at low PSA values132.

At the moment, to obtain a patient distinction in one of these groups as best as possible, therefore reciving different treatment regimens, the most promising imaging procedure are represented by DWI/DCE-MRI (for locoregional assessment)133,134 and by PET/CT with radiolabelled choline for extraregional restaging135.

The use of DWI (Diffusion Weighted Imaging) allows in vivo measurement of diffusion coefficients of biological tissues, taking advantage from qualitative assessment of relative tissue signal attenuation136-139 and quantitative analysis140-141, using parameters like apparent ADC142. .Images can be acquired quickly without the administration of exogenous contrast medium143, showing also good performance in metastatic lymph nodes detection144-148.

DCE-MRI, basing on repetitive acquisition of sequential images during the

passage of a contrast agent within a tissue of interest149,150, enables noninvasive visualization of tissue vascularity,tissue cellularity and cell

24

membrane integrity151-156. This features allow tumor characterization, thus malignancy has peculiar vascular features (chaotic structure, arteriovenous shunting, high permeability, and areas of hemorrhage157-158). Some quantitative parameters have been developed to estimate pharmacokinetic properties, such as K trans , v e , and kep159-160 (Fig.4).

Figure 4 Application of multiparametric MRI techniques in a patient with PCa A) T2-weighted, B) diffusion

weighted (DWI), C) apparent diffusion coefficient (ADC) and D) dynamic contrast enhanced (DCE) imagesThe tumor appears dark on the axial T2-weighted image (arrow); the corresponding area shows restricted diffusion on the DWI and ADC images as well as abnormal contrast enhancement on the DCE

axial image (as evident from abnormal red color coding of the tumor)

PET/CT scanning has brought about a revolution in the imaging of many

common cancers, being incorporated into routine radiation therapy planning for its intrinsic capability to assess methabolic status of lesions and to estimate extension of disease, thereby defining the target volume.

25

The leading PET tracer, [18F]-FDG presents limitations in imaging PCa: Although it may accumulate in aggressive and undifferentiated tumours, PCa often presents with poor avidity for it161.

Choline, labelled with either [18F]- and [11C] have been proposed as PET tracers to study PCa patients, presenting similar results162,163.

Choline is an essential component of phospholipids of the cell membrane, and cell proliferation and upregulation of choline kinase seems to be mechanisms suggested for the increased uptake of this tracer in PCa164. The presence of choline transporters also seems to be involved in the process of its uptake and trapping into cancer cells165. In patients with PCa recurrence choline-PET/CT have been particularly investigated166,167, showing great results in terms of accuracy for restaging patients with distant metastases and nodal involvement (Fig.5).

Figure 5 [18F]-FECH PET/CT shows a pathological uptake pertaining to left iliac nodes in a patient with PCa with biochemical relapse (PSA=1,7 ng/dl)

26

A recent methanalys168 upon restaging patients with biochemical failure only after local therapy with curative intent by using choline PET or PET/CT (Tab.

6) asserts that “the pooled results fulfilled the criteria for high diagnostic evidence in the classification described by Jaeschke169 and coworkers”.

YEAR AUTHOR PTS,n= SENSITIVITY SPECIFICITY

2003 De Jong 36 55 100 2003 Picchio 100 84 94 2007 Vees 11 _{38 33} 2007 Rinnab 50 95 40 2007 Scattoni 25 100 67 2008 Pelosi 56 83 96 2008 Reske 49 70 75 2009 Rinnab 41 93 36 2009 Winter 8 100 86 2010 Breeuwsa 70 _{81 100} 2010 Fuccio 25 86 100 2010 Giovacchini 358 85 93 2010 Giovacchini 170 87 89 2010 Winter 6 100 67 2011 Bertagna 45 60 91 2011 Langsteger 40 91 89 2011 McCarthy 26 _{96 96} 2012 Panebianco 55 92 33 2012 Picchio 78 89 100

Table 6 Results of Choline PET/CT for PCa in terms of sensitivity and specificityy by a recent methanalys (Umbehr et al,Eur Urol 2013168)

27

AIM OF THE STUDY

Focus of our project was recurrent PCa after radical local treatment. We aimed to investigate the impact of f-MRI and [18F]Fluoroethylcholine ([18F]FECH) PET/CT in treatment-decision making in patients with PCa recurrence after a curative first-line treatment, demonstrating that an accurate pre-treatment restaging using the combination of these new functional imaging modalities may be used to optimize treatment decision making in patients with recurrent PCa, improving final outcome.

Imaging techniques were used to identify sites of relapse, define different pattern of recurrence (loco-regional or distant metastasis), determine the rate of patients presenting recurrence in form of oligometastatic disease (who will potential benefit from Hazard Ratio Image-Guided Radiation Therapy (HR IGRT) or in form of plurimetastatic disease candidate for ADT).

The possibility to guide therapeutic strategies based on the pattern of the recurrence leaded to introduce in the setting of PCa new treatment options such as HD IGRT in the case of oligometastatic disease, deserving ADT. In fact, in presence of regional and distant metastasis we tailored treatment based on the pattern of recurrence that we would demonstrate: patients with 1- 5 lesions (oligometastatic disease) would have treated with HD IGRT while patients with > 5 lesions (plurimetastatic patients) would have undergone ADT.

28

As primary end points, our study intended:

 to create a new diagnostic algorythm of recurrent prostate cancer patients with the scope of accurate definition of the pattern of disease relapse (oligometastatic vs plurimetastic patients);

 to determine the impact of combined DWI (Diffusion-Weighted Imaging), DCE (Dynamic-Contrast Enhanced) MRI and [18F]FECH PET/CT for patients’ restaigind as compared to each single procedure;

 to assess in an observational way how our results could modify clinical practice and patient management;

 to observe how an accurate patients’ restaing could impact on reatment outcome.

As secondary end points, we aimed:

- to introduce HD IGRT as treatment option in presence of oligometastatic disease to investigate if this treatment modality may be routinely offered in the early phase of disease relapse.

29

STUDY DESIGN

Our study was a prospective non randomized phase I trial in patients with relapse of PCa after first line curative-intent treatment. Patients entered the study during a period of 18 months. The patients participation time was defined as the length of time between enrollment of the last patient into the study and the end of the study (observation period) and was set to 6 months.

Inclusion criteria:

-Patients of all ages who have previously undergone treatment with radical intent;

-A global good performance status (PS) was mandatory for patients enrollment;

-Patients with every T and N stage will be enrolled; -Every Gleason score was accepted;

-Patients with cancer relapse documented by two consecutive rises, with PSA levels > 1.5 ng/ml.

Exclusion criteria:

-Patients not treated with radical intent;

-Patients having hormonal therapy at the moment of biochemical relapse; -Patients treated for primary cancer but with a biochemical relapse < 1.5 ng/ml;

30

-Patients with low PS and low life expectancy.

STUDY PROCEDURES: After a screening visit, patients underwent two

consecutive imaging procedures: f-MRI and [18F]FECH PET/CT. Results of those two combined imaging modalities were used to classify patients in the following groups: patients with loco-regional relapse, patients with oligo-metastatic recurrence and patients with oligo-metastatic disease. Accordingly, the classification derived from the combination of f-MRI and [18F]FECH PET/CT have guided the selection of the most suitable treatment defined as follow:

a) patients with local relapse (as lodge relapse) candidate for standard salvage radiation therapy ( usually as Intensity Modulated Radiation therapy, IMRT) with MR guided treatment planning

b) patients with oligometastatic disease furher divided into:

 patients with loco-regional relapse (as lodge or nodes disease) that were candidate for EBRT (External Beam Radiotherapy) with combined MRI and [18F] FECH PET/CT guided treatment planning, possibly followed by ADT;

 patients with not strictly regional oligo-metastases recurrence (included bone localization) candidate for stereotactic HD IGRT (in case of contraindication or according to clinician, they would have received ADT);

31

STATISTICAL DATA ANALYSIS: Descriptive statistics (n, mean, standard

deviation, median, minimum and maximum) were calculated for quantitative variables; frequency counts by category were given for qualitative variables. Confidence intervals were given where appropriate. Prognostic value of DW/DCE MRI and [18F]FECH PET/CT were evaluated in relations to overall survival (OS), disease free survival (DFS) at 1 and 2 years and using clinical observation as gold standard, calculated on the basis of normal approximation If not else wise stated, these intervals were two-sided and provide 95% confidence.Mc Nemar test was used to compare differences of the two imaging procedures in terms of statistical significance. Statistical analysis was performed using SAS System.

PATIENTS FOLLOW-UP:

Patients were followed with a frequency of every 3 months from their enrollment, up to two year after. The following check/procedures were performed at each follow-up visit and at the end of treatment period: 1) total PSA; 2) physical examination and vital signs and body weight; 3) concomitant medication; 4) assessment of adverse events; 5) ECOG performance status.

32

PATIENTS AND METHOD

The protocol had the initial approval of Ethics Committee of the University Hospital of Pisa.

PATIENTS POPULATION:

From June 2011 to June 2013 we enrolled 30 consecutive patients with PCa who have already undergone therapy with radical intent, presenting biochemical relapse of disease.

All patients fullfilled all the inclusion criteria previously described.

At the time of enrolment, patients mean age was 72 years ±7 years, range 56-83. The mean age when they were initially treated was 66 years ±6 years, range 54-78; therefore biochemical relapse occurred after a median of 6 years from primary treatment.

The most representative Gleason score founded by biopsies was 7, resulting from 3+4 score in 26 % and from 4+3 in 33% of our patients population. The most representative “T” staging at post-surgical treatment was “T2” (63% totally). Regarding “N” parameter, only 5 patients were classified as “N1” (11%), the others were N0. No “M”+ patients were enrolled.

Patient were initially treated as follow:

 23 underwent RP;

 3 underwent RT;

33

No patient were on ADT at time of enrollment, but 46% patients (n=11) underwent previous ADT for an average of 25 months. ADT withdrawal of at least 1 year was guaranteed.

The average total PSA at the moment of their enrollment was 3.3 ng/ml, ranging from 1.5 to 13.05. PSAdt values ranged between 3 and 20 months.

Table 6. recaps patients population.

Pts Enrolled Mean Age At Diagnosis

Mean Age At

Relapse Average PSA

30 66 72 3,3 Previous Therapy RP RT RP+RT ADT 23 3 4 11 Gleason Score GS=6 GS=7 GS=8 GS=9 3 18 5 4 Stage At Diagnosis T1cN0 T2aN0 T2bN0 T2cN0 T3aN0 T3bN0 T3bN1 1 1 5 14 2 4 3

34

METHODS:

After a screening visit, sign of inform consent and enrolment, patients underwent two consecutive imaging procedures within a week: DWI, DCE-MRI and [18F]Flurocholine PET/CT before treatment planning, for a total of 60 exams performed (30 MRI and 30 PET).

DWI, DCE-MRI: All studies (n°=30) were performed at Department of

Diagnostic and Interventional Radiology at Santa Chiara Hospital, Pisa. MRI included imaging of abdomen and pelvis (with dedicated coils at 1.5 T MR) and prostate (with dedicated endorectal coil on a 3T MR scanner, (GE Signa Excite HD 12.0 Twin Speed 8-channel scanner, General Electric, Milwaukee, WI) and at 1.5T scanner (GE Signa Excite HD 12.0 Echospeed 8-channel scanner, General Electric, Milwaukee, WI).). The protocol included two different DWI sequences with multiple b-values (low and high b values) used to assess the tumor apparent diffusion coefficient and perfusion. The low b-values were in the range of 0-100 s/mm2, the high b- values were in the range of 800-1200 s/mm2. The MRI of the abdomen and pelvis protocol included the following sequences: 1) T1-weighted IN-and-OUT of phase; 2) fat suppressed T2-weighted; 3) breath-hold DWI; 4) free breathing DWI; 5) contrast enhanced fat suppressed T1-weighted.

The MRI of the prostate will include the following sequences: 1) T1-weighted imaging; 2) fat suppressed T2-weighted; 3) free breathing DW imaging; 4) DCE MRI.

35

The sequence parameters were optimized for the 1.5T and 3T MR scanners, respectively. The sequences were designed at the scanner in order to maximize the single MR scanner imaging performances. The study focused on prostate was performed with endorectal coil (8-ch 1.5 T ID Medrad P/N

M64ERA8-HD eCoil™ or 8-ch 3.0T ID MEDRAD P/N M128ERA8-HD eCoil™).

Data interpretation and analysis: Two radiologists independently reviewed the images to determine tumor location. The same two radiologists measured the ADC (apparent diffusion coefficient), the ktrans (the transfer constant), the kep (the rate constant) and ve (the fractional volume) of the lesion suspicious for cancer recurrence. The post processing included mono- and bi-exponential ADC map calculation for the abdomen and pelvis and prostate imaging. For the prostate imaging DWI was fused with the T2 weighted images. Each MRI scan was classified as follow: negative/ positive/ indeterminate. Results were used to classify patients in the following groups: patients with local relapse, patients with oligo-metastatic recurrence and patients with metastatic disease.

[18F]FECH PET/CT: All studies (n°=30) were performed at Nuclear

Medicine Institutes at Santa Chiara Hospital in Pisa (n=10), at USL 1 Hospital in Massa (n=2) and at Istituto Radiologico Toscano in Pistoia (n=18), with an integrated PET/CT system (respectively Discovery 710; GE

36

Philips Healthcare, Best, The Netherlands). [18F]-FECH (IASOcholine® from IASON, Graz, Austria) was injected i.v. (about 4 MBq/kilogram of body weight). As a routine protocol, imaging started 1 minute after intravenous injection, with acquisition of dynamic PET images at one constant bed position of the pelvic region (covering the pelvic floor and urinary bladder) for 6 minutes (1 minute per frame) to overcome the effect of urinary activity in the bladder.

After this early phase patients rested for approximately one hour. The whole body acquisition was performed thereafter with the 3-dimensional mode, 2 min per bed position from base of the skull to midthigh (six- or seven-bed-position). Images were reconstructed with standard reconstruction ordered-subset expectation maximization iterative algorithm (two iterative steps) and reformatted into transverse, coronal, and sagittal views. Unenhanced CT were performed for localization and attenuation correction (120 kV, 0.5 second per rotation, 3.27-mm reconstructed section thickness). CT features are summarized in Table 8.

37

CT PARAMETERS FOR [18F]FECH PET/CT

Marker Anterior-Superior Iliac Spine Scout S50.0 – I 250.0 90°

120 kV – 10 mA

CT Helical 3.75 mm – pitch 0.984 – 0.5s/rot spacing 3.27 mm – 47 slice 120 kV - AutomA™ - SmartmA™ 15-100 mA Noise Index 25.0

Recon 1: PET AC WideView DFOV 70.0 mm Recon 2: Stnd DFOV 50.0 mm

PET Static List Mode 1 bed – 00:06:00

VUE Point FX™ - SharpIR™ - 32 subset 3 iter Gauss 6.0 mm – 256x256 PET Dynamic Unlist 1 bed – 6 x 00:01:00

VUE Point FX™ - SharpIR™ - 24 subset 3 iter Gauss 5.0 mm – 128x128 Marker OM

Scout S0.0 – I1100.00

CT Helical 3.75 mm – pitch 0.984 – 0.5s/rot spacing 3.27 mm 120 kV - AutomA™ - SmartmA™ 15-100 mA Noise Index 25.0 Recon 1: PET AC WideView DFOV 70.0 mm

Recon 2: Stnd DFOV 50.0 mm PET 00:02:30/bed

VUE Point FX™ - SharpIR™ - 24 subset 3 iter Gauss 5.0 mm – 256x256

Table 8 CT parameters applied in PET/CT with [18F]FECH for dynamic and total-body scan

Data interpretation and analysis: PET/CT images were interpreted by two nuclear medicine specialists, aware of the patients medical history and PSA values. Images were read sequentially using advanced PET/CT review software (Advantage for Windows, versions 4.2 through 7; GE Medical Systems), which allows simultaneous scrolling through the corresponding PET, CT, and fusion images in transverse, coronal and sagittal planes. Semiquantitative analysis of the abnormal radiotracer uptake was performed by using the maximum standardized uptake value (SUVmax).

Each [18F]FECH PET/CT scan was classified as follow:

38

following groups: patients with local relapse, patients with oligo-metastatic recurrence and patients with metastatic disease.

Comparative MRI and [18F]FECH PET/CT: Results of MRI and [18F]FECH

PET/CT were combined to determine the rate of agreement. Furthermore results of the two imaging modalities were combined to classify patients in the following groups: patients with local relapse, patients with oligo-metastatic recurrence (≤5 total lesions detected, either as locoregional disease or not) and patients with (pluri)metastatic disease (≥5 total lesions detected). Patients were then treated according to the standard of care of our center, that is:

a) patients with local relapse: salvage EBRT with MRI guided treatment planning;

b) patients with oligometastatic disease, furher divided into:

- loco-regional relapse: EBRT with combined DWI/DCE MRI and [18F]FECH PET/CT guided treatment planning (possibly followed by ADT);

- oligo-metastases recurrence (included “extraregional” localization, as bone): stereotactic HD IGRT (in case of contraindication or according to clinician, suitable for ADT);

- plurimetastatic disease: ADT.

Rate of changes in final classification of patients were determined as well as for each specific sub-cathegories.

39

Changes in the intention-to-treat after the sole f-MRI, the sole [18F]FECH PET/CT and after the combination of the two imaging modalities was also determined.

HD IGRT and ADT were administered according to the standard protocol.

DATA COLLECTION:

Data were recorded on an apposite Case Report Form (CRF).

STATISTICAL DATA ANALYSIS:

Descriptive statistics (n, mean, standard deviation, median, minimum and maximum) were calculated for quantitative variables; frequency counts by category were given for qualitative variables. Confidence intervals were given where appropriate. If not else wise stated, these intervals were two-sided and provide 95% confidence. Intervalsf or observed specificity, sensitivity and accuracy of f-MRI and [18F]FECH PET/CT using clinical observation as gold standard, will be calculated on the basis of normal approximation. Prognostic value of DW/DCE MRI and [18F]FECH PET/CT derived parameters were evaluated in relations to overall survival (OS) and disease free survival (DFS) at 1 and 2 years. Mc Nemar test was used to compare differences of the two imaging procedures in terms of statistical significance. Statistical analysis was performed using SAS System.

PATIENTS FOLLOW-UP:

Organized as previously explained. Five patients, enrolled at the end of 2013 and during the year 2014, didn’t conclude their follow up.

40

RESULTS

DWI/DCE -MRI RESULTS: MRI was positive in 20 patients, negative in 8 patients, indeterminate in 2 patients.

Positive scans (n=20) were further classified as:

 isolate prostatic lodge disease (=11)

 oligometastatic (≤ 5 sites) disease (=9, further classified as oligometastatic with loco-regional disease, n=8 and oligometastatic with extraregional disease, n=1)

 plurimetastatic (>5 sites) disease (=Ø).

Indeterminate results were found in case of lodge (=1) and locoregional lymphnodes (=1).

[18F]FECH PET/CT RESULTS:

PET was positive in 17 patients, negative in 10 patients, indeterminate in 3 patients.

Positive scans (n=17) were further classified as:

 isolate prostatic lodge disease (=3);

 oligometastatic (≤ 5 sites) disease (=12, further classified as oligometastatic with loco-regional disease, n= 3 and oligometastatic with extraregional disease, n=9);

41

Indeterminate results were found in case of lodge (=1) and bone involvement (=2).

MATCHED RESULTSANDAGREEMENT:

Comparing the results of DWI/DCE MRI and PET/CT:

 n=11 positive at both MRI and PET scans;

 n=3 negative at both DWI/DCE MRI and PET/CT;

 n=3 positive at DWI/DCE MRI and indeterminate PET/CT;

 n=6 positive at DWI/DCE MRI and negative at PET/CT;

 n=1 indeterminate at DWI/DCE MRI and negative at PET/CT;

 n=1 indeterminate at DWI/DCE MRI and positive at PET/CT;

 n=5 negative at DWI/DCE MRI and positive at PET/CT.

Tables 9 and 10 summarize the results.

[18F]FECH PET/CT DWI/DCE -MRI POSITIVE 17 20 Lodge 3 11 Oligometastatic 12 10 Plurimetastatic 2 0 NEGATIVE 10 8 INDETERMINATE 3 2

Table 9 Individual results for each of 30 patients taking account of [18F]FECH and DWI/DCE MRI separately

42

Table 10 Matched patient population results for [18F]FECH and DWI/DCE MRI. POS=positive results; NEG=negative results; IND=indeterminate results

McNemar test was used in a case-by case analysis to match the positive and negative findings of the all imaging procedures to evaluate the concordance rate (Table 11).

Table 11 Comparison beetween total positive and negative cases for both [18F]FCH-PET and DWI/DCE MRI. The McNemar Test result shows a p=1, which is considered not statistical significant. Numbers of patients considered was 25 for the absence of those with indeterminate findings.

PTS, n= COMPARED RESULTS

11 POS MRI /POS PET/CT

3 NEG MRI/ NEG PET/CT

3 POS MRI/ IND PET/CT

6 POS MRI/ NEG PET/CT

1 IND MRI/ NEG PET/CT

1 IND MRI/ POS PET/CT

5 NEG MRI/ POS PET/CT

DWI/DCE–MRI

+

-

+

-

43

Moreover we match result relatively T” (as lodge disease, Table 12), ”N” (Table 13) and “M” (Table 14) parameters to find if the two methods were concordant or not.

p=0.0026

Table 12 Comparison beetween positive and negative cases regarding “T parameter” (considered as lodge)

for both [18F]FCH and DWI/DCE MRI . With McNemar Test we obtained a p=0.0026 (statistical significant). Patients included were 28 (we didn’t consider 2 lodge uncertain findings). Lodge disease was

considered even when included for “oligo- or pluri-metastatic disease”

DWI/DCE-MRI

+

-

+

-

Table 13 Comparison beetween total positive and negative cases regarding “N parameter” for both [18F]FCH and DWI/DCE MRI . The McNemar Test result shows a p=0.4497, which is not considered

statistical significant. The total patients were 29, because we didn’t considered the one with uncertain findings at lymphnodes. All “N” results were considered, both for oligo-and pluri-metastatic disease

DWI/DCE–MRI

+

-

+

-

44 DWI/DCE–MRI + - [18 F]FECH-PET/TC + 1 7 p=0.0233 - 0 20

Table 14 Comparison beetween total positive and negative cases regarding “M parameter” for both

[18F]FCH and DWI/DCE MRI . The McNemar Test result shows a p=0.0233, which is considered statistical significant. All “M” results were considered, both for oligo-and pluri-metastatic disease. The total patients were 28, because we didn’t considered the two patients with uncertain findings at bones

FINAL RESULTS AND STAGING:

Combining both DWI/DCE-MRI and [18F]FECH PET/CT results 27 patients (90%) were eventually diagnosed with recurrent disease and 3 patients (10%) were negative. No patient had indeterminate results.

In particular, 7/10 patients that were negative at PET/CT and 3/3 patients that had unconclusive findings at PET/CT were classified as positive by adding DWI/DCE-MRI findings. Conversely, 5/8 patients (62%) that were negative at DWI/DCE-MRI imaging and 2/2 (100%) patients with unconclusive findings at DWI/DCE-MRI turned out to be positive combining the PET/CT findings (Fig.6).

Negative results were true negative in 3/8 cases (37%) for DWI/DCE MRI and 3/10 (30%) for PET/CT, respectively.

45

Figure 6 A representation of the separate contribuites of each imaging modality to reach the final

classification as “positive”

Considering both DWI/DCE-MRI and [18F]FECH PET/CT results, the final staging was (Table 15):

 patients with local relapse, n=7;

 patients with oligo-metastatic disease (≤5 total lesions), n=18 further classified in with locoregional disease (n=8) and oligometastathic disease including locoregional and distant sites (n=10);

46 FINAL RESULTS POSITIVE 27 Lodge 7 Oligometastatic 18 Plurimetastatic 2 NEGATIVE 3 INDETERMINATE 0

Table 15 Final results in terms of disease classification in our patients population

Therefore, based on the [18F]FECH PET/CT-derived classification final

staging was changed in 10/30 patients (about 33%), as follow: 4/10 cases

from negative/indeterminate to local relapse; 5/10 from

negative/indeterminate to oligo-metastatic disease (loco-regional spread). In the remaining patient an indeterminate finding became oligometastatic disease with extraregional spread.

DWI-DCE MRI-derived staging changed in 15/30 patients (50 %).

Indeterminate/negative DWI-DCE MRI were transformed in local relapse (n=1), oligometastatic disease with loco-regional spread (n=2) and extra-regional disease (n=3) and plurimetastatic disease (n=1). Patients with local relapse were ri-staged as oligometastatic with loco-regional disease (n=1), as oligometastatic with extra-regional disease (n=4). Cases of oligometastatic

47

loco-regional disease were transformed in oligometastatic extra-regional disease in 2 patients and plurimetastatic disease in the remaining one.

Fig.7 shows in a syntetic way how the final classification was reached.

Therefore, by the combination of each imaging methodology, we obtained an overall change in 25/30 patients (83%).

Figure 7 A representation of the final staging obtained by matching [18F]FECH PET/CT and DWI/DCE-MRI separate findings

CHANGE IN “INTENTION TO TREAT”:

Combining both DWI/DCE-MRI and [18F]FECH PET/CT results patients had a change in their final treatment as compared to the initial intention to treat in 25/30 cases.

48

Seventeen patients with negative imaging that would have been either follow-up without any treatment or treated empirically with ADT based on PSA level, had change in treatment as follow:

 salvage radiotherapy with IMRT (local relapse, n=5);

 IMRT on the lodge and pelvic EBRT (oligometastatic loco-regional disease, n=7);

 HR IGRT for the detection of oligometastic extra-regional disease (n=4);

 ADT for plurimetastatic disease (n=1).

Five patients with local relapse candidate for salvage IMRT changed treatment plan in: combination of IMRT and pelvic EBRT (oligometastatic loco-regional disease, n=1) and HD IGRT for oligometastic extra-regional disease, n=4 .

The patients candidate to combine IMRT and EBRT (oligometastatic loco-regional disease, n=3) were then candidate to HD IGRT (xtraloco-regional oligometastatic disease, n=2) or ADT ( plurimetastatic disease, n=1).

Therefore the new intention to treat is standard salvage IMRT in 7 patients, EBRT or HD IGRT, possibly followed by ADT in 18 cases and just in the 2 cases with plurimetastatic disease ADT.

Since the study is observational, results of the combined imaging modalities were not used to change the real patients treatment that was based on the referring physician clinical decision and routine practice.

49

Figure 8 A representation of the changing in “intention to treat” obtained by matching [18F]FECH PET/CT and DWI/DCE-MRI separate findings

The final treatment of the patients was as follow:

 salvage IMRT in the 7 patients with local relapse;

 pelvic irradiation (n=4), pelvic + nodal irradiation with HD IGRT (n=4) in presence of with loco-regional disease;

 lodge IMRT and HD IGRT on bone (n=4); HD IGRT on bone (n=3), a combination HD IGRT on bone and ADT (n=3) in presence of local and distant disease or distant disease;

 ADT in the 2 patients with plurimetastatic disease.

The three negative patients underwent respectively 1 ADT and 2 watchful waiting.

50 Table 16 summarises treatment data.

PATIENTS FINAL STAGING AND RELATIVE TREATMENT

LOCAL RELAPSE 7 Salvage RT n=7

OLIGOMETASTATIC

DISEASE 18

Pelvic Irradiation n=4 Pelvic and Nodal Irradiation n=4

Bone HD IGRT n=3 ADT + Bone HD IGRT n=3 Bone and Lodge HD IGRT n=4 PLURIMETASTATIC

DISEASE 2 ADT N=2

Table 16 A summary of patients final staging and their relative treatment

RESULTS AND PSA VALUE: [18F]FECH PET/CT and MRI scan were negative at any PSA value (ranging from 1,5 ng/dl to 13 ng/dl), without significant correlation (Fig.9 and 10).

Figure 9 Match of Positive/Negative [18F]FECH PET/CT with PSA value 0 2 4 6 8 10 12 14 16 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 PS A val u e

[

_{]FFECH PET/CT and PSA}

POSITIVE PET NEGATIVE PET

51

Figure 10 Match of Positive/DWI/DCE MRI with PSA value

TREATMENT RESPONSE: No modification of PS and no adverse event

were recorded troughtout the study. Treatment response based on clinical and PSA data at June 2014 is represented in Fig.11.

Figure 11 Results of follow-up for 25 patients in terms of recovery, stability and progression disease

0 2 4 6 8 10 12 14 16 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 PS A val u e

DWI/DCE MRI and PSA

NEGATIVE MRI POSITIVE MRI 0 5 10 15 20 25 90 180 270 360 450 540 630 730

Treatment response

52

Figure 12 fMRI detection (DWI+ spectroscopy, A) of peri-rectal bilateral relapse in a patient with biochemical recurrence (1,61 ng/ml) after RP [18F]FECHPET/CT show no lesions with pathological uptake,

particularly at lodge level (B). Radiation salvage therapy planning was therefore developed on the basis of MRI findings (C). Patients eventually underwent lodge RT with VIMAT technique (Volumetric

53

Figure 13 [18F]FECH PET/CT transaxials slices (CT and fused PET/CT images, respectively) show [18F]FECH uptake at paraortic lymphnodes (A). At the moment of the scan, patient present a biochemical

relapse of PCA, with PSA= 5,68. He was classified as oligometastatic, for presenting exclusive nodal involvement (<5). Radiation therapy planning (B) was based on PET/CT exam, and patient underwent HD

54

Figure 14 A case of PCa biochemical relapse (PSA= 1,73 ng/ml) 1 year after RP. [18F]FECH fused PET/CT respectively coronals (A, on the left ) and transaxial (B, on the left ) images show area with abnormal uptake interesting left acetabulum, corresponding at CT images (A-B,on the right) at osteoaddensant area. fMRI in the same patient had negative findings. Patient, classified as oligometastatic, underwent HD IGRT on the basis of [18F]FECH PET/CT imaging (24 Gy, as standard protocol).

55

Figure 15 A patient presenting with PSA= 4,2 ng/ml, 3 years after RP. Patient, initially classified as with lodge relapse according to DWI/DCE MRI findings (A) and therefore suitable for salvage therapy , became plurimetastatic according to [18F]FECH PET/CT findings. [18F]FECH PET/CT showed multiple disease localizations involving respectively bone (B), lodge (C), pelvic (D) and mediastinal nodes (E), thus patient underwent immediate ADT

56

DISCUSSION

Despite improvement in both staging and treatment of patients with PCa, biochemical recurrence occurs in high percentage of patients within 10 years after curative intent therapy, often requiring second-line treatment.

When PSA rises, definition of disease recurrence location and extent is fundamental. This is exactly what clinicians expect when requiring an imaging procedure, thus helping guiding the select of the most appropriate treatment.

Our study aimed to evaluate the impact of the combination of DWI/DCE-MRI and [18F]FECH-PET/CT in patients with recurrent PCa for the accurate definition of the pattern of disease relapse and the subsequent selection of the treatment. In fact, a substantial number of patients with biochemical relapse undergo ADT despite the absence of any pathological or radiological demonstration of disease (PSA sine materia). However, long-term maintenance of ADT is a common practice that does not affect survival120 while presenting a wide spectrum of side effects170.

Before the era of high-resolution imaging tecniques, patients with PSA rising after first line curative treatment, underwent either ADT or watchful waiting, depending on PSA levels or presence of symptoms. Otherwise, when isolated local recurrence either limited to the lodge or with locoregional lymph-nodes involvement are demonstrated, the best therapeutic option is radiation